Assembly Standing Committee on Privacy and Consumer Protection

Good afternoon. We will call this hearing of the Privacy and Consumer Protection Committee in order, and we will be doing our bills today in file order. And so I'm going to express immense gratitude to Miss Quirk-Silva, who is here, here on time and ready. So, to all of our colleagues, that's. How you get kudos in this Committee. So, with that, Miss Quirk-Silva, we are going to start as a Subcommitee. Hopefully more Members will join us shortly. But if you want to begin, the floor is yours when you're ready. And this is AB 1788.

Thank you, Madam Chair. And Members, I first want to say thank you to the Committee staff for diligently working with our office during this busy time of year. And I accept the amendments. Today, I present Assembly Bill 1788, which allows counties to establish mental health multidisciplinary, Disciplinarian teams focused on addressing the mental health needs of justice involved individuals who hold the values of confidentiality. Over the last decade, California's correctional facilities have experienced a massive surge in mental health cases amongst their population.

The prevalence of addressing mental health needs among justice involved individuals has soared from 20% in 2010 to an outstanding 53% in 2023. To fulfill California's commitment to rehabilitation and reduce recidivism, we need to establish efficient practices to assist individuals dealing with mental health difficulties within the justice system. However, procedural barriers have hampered our efforts, including the lack of authorization and current law to facilitate the sharing of critical information between various entities.

AB 1788 grants counties the authority to create multidisciplinarian teams to expedite the assessment of mental health services by allowing the appropriate sharing of confidential information. It also includes safeguards to protect confidentiality, ensuring that shared information is used solely to provide services with no implications for enforcement or criminal investigation. By prioritizing mental health wellbeing and improving communication channels, we can create a foundation for targeted care, ensuring effective rehabilitation, adequate mental health treatment, and reducing the likelihood of recidivism within our justice involved community.

With me today to provide testimony and answer any questions from the Committee and the sponsor of AB 1788 is Mister Jeff Pupkett, assistant sheriff of Orange County Sheriff's Department.

Thank you.

Good morning, Committee Members. My name is Jeff Puckett and I'm an assistant sheriff with the Orange County Sheriff's Department. I currently oversee our custody operations command and prior to that, oversaw both our behavioral health and homeless teams in the county. I'd like to thank the Committee for Hearing us today regarding AB 1788 and state that we are proud to partner with Assemblywoman Quirk Silva on this Bill.

As many of you know, 2020 saw the passage of Assembly Bill 210, which allowed for increased information sharing between multidisciplinary teams to assist those experiencing homelessness. Some of the information shared included health information, mental health information, criminal history, and social services information. I think it's important to note that law enforcement was not a recipient of mental or medical health information, which is still carefully protected by HIPAA.

However, being able to share criminal history information with providers allowed for better shelter placement and a higher level of care. Assembly Bill 210 allowed for three expected outcomes to be achieved, which include number one, a better quality of service delivery, two, increased ability to match better services to the client and three, reduced costs by decreasing duplication of services. It's my opinion that the successful implementation of Assembly Bill 210, coupled with the mdts that we had in place, are what allowed Orange County to decrease homelessness.

According to the 2020 and 2022.0 in time counts, post COVID correctional facilities have seen an increase in mental health population, with Orange County in particular having jumped from 22% to 45%. For those with an open mental health, these numbers are even more striking considering the average daily population has decreased by 50%. We've also seen a sharp increase in those experiencing substance abuse disorder.

In Orange County alone, approximately one third or 900 inmates of our average daily population are receiving medically assisted treatment to help them overcome opioid addiction. Like many sheriff's departments, the Orange County Sheriff's Department has taken numerous proactive steps to assist the jail population experiencing both behavioral health issues and substance abuse disorder.

Some of these steps that we've taken include an increase in behavioral health, housing units, an increase in LP's beds, crisis intervention training for all staff, expansion of the medically assisted treatment program, new reentry programs, and a reduction in late night releases. AB 1788 is the next proactive step to strengthen and allow for broader communication to better serve our jail community.

As the Legislature has done with the past and with other community concerns, including AB 210, SB 1342, AB 271, and welfare and Institutions Code 18951, which all allowed for increased information sharing between mdts. We know that mdts work, I've seen this firsthand, but we know that they only work if we can communicate. All the pieces of prior legislation allowed for information sharing and communication to better serve the community. I would respectfully ask the Committee Members for an aye vote so that we can continue this as it pertains to behavioral health within the state.

Thank you so much. Anybody else here in support of AB 1788? Thank you.

Thank you.

Clifton Wilson. On behalf of the California Faculty Association and the California Academy of Child and Adolescent Psychiatry and support, and then also on behalf of the California State Association of psychiatrists with the position of support and seek amendments. Thank you.

Thank you. The mic is geared to the height of the chair, I think. Anyone here in opposition to this Bill? Yeah, please come on up. We will give you three minutes as well, when you're ready.

Sure. And to clarify, our position is opposed unless amended.

Thank you.

Good afternoon, chair and Members. Tracy Rosenberg from Oakland Privacy. We are a regional coalition that advocates for guardrails and safeguards in the interests of privacy protection, civil rights and community consent. First, we want to thank the Committee and the author for the commitments which address the vast majority of our concerns, which are about sharing confidential medical information about vulnerable people. We just briefly wanted to mention the one or two concerns that remain.

We would like to see the Bill define the agencies that are authorized to participate in these information sharing terms. Teams, concretely, without a not limited to disclaimer, it's sort of a foundational thing in privacy regulation that the who is sharing information with who needs to be concrete and specific. We think that with three and a half months left of potential commendment time, there should be some way to survey counties to see if any agencies need to be added that are not currently listed in the Bill.

And we would recommend taking that approach instead of the not limited to a disclaimer. Secondly, we're a little cautious about the role of law enforcement in these task forces when people are no longer incarcerated. Traditionally, departments of parole and probation have operated separately from police departments for a reason. And it's usually pretty destabilizing for people that have been released from carceration to sort of have police continuing to be involved in their cases and their care destabilizing. And we think it can be triggering.

We have always heard that police don't want to be social workers. We agree, and we suggest that maybe probation and parole, at least when people are no longer incarcerated, would be the more appropriate participants in these task forces. That's all, and thank you for hearing our concerns.

Perfect. Thank you. We're going to establish a quorum before one of my colleagues walks out. Call to roll.

[Roll Call]

We have a quorum. Thank you. And with that, any other witnesses in opposition to this Bill seeing none. We will bring it back to the Committee. Any questions or comments? Committee Members seeing none. I will just add, I want to thank you for taking the amendments as well. I think they do go far in protecting privacy, which is obviously the role of this Committee. But I do want to give a huge shout out to Orange County.

Your coordination really has proven to be successful as it relates to your number of unhoused. And so I think, you know, you're bringing some of that ability for us to coordinate better in serving the people who need us most. And as we know, our mental health crisis is very, very real, and the individuals in our community who are suffering need our support.

And so the ability to do that in a more seamless and better way, which I know our counties are working incredibly hard to do, is really important work. So I want to thank the author for her leadership in that space. And with that, would you like to close?

I would. I would just end or close with. The data shows that more and more of those that are entering our county jails are suffering with some type of mental health issue and they're staying longer in the county jails. They used to be somewhat short stays and now they can be 1234 years. And so they are already interfacing with different organizations in order to help get the treatment they need.

But when they leave, being able to continue that coordinated effort to support them, I think gives them the best chance of moving on into a healthy next step. So with that, I would ask for an aye vote.

Thank you. With that, do we have a motion on this Bill? Motion and a second. Madam Secretary, please call the roll.

[Roll Call]

That Bill has six votes, so it's out. But we will leave the roll open for absent Members. Thank you, Miss Quirk-Silva.

Thank you, Members. We have a motion on the consent calendar. We have a motion. We have a second on the consent calendar. Second. So there are two items on the consent calendar. AB 2652 by Semblance Marisucci. The motion is do pass to the Appropriations Committee with a recommendation to consent calendar. Nab 2885. Bauer Cahan with a motion to do pass to the Appropriations Committee with recommendation of the consent calendar. Please call the roll.

[Roll Call]

The consent calendar has six votes, it is out. But we will leave the roll open for absent Members. With that, we will. We have no authors here besides Committee Members, so authors in the ether that are listening. You might want to head over. But we will move on to miss Wicks, because she is here, prepared and ready. So we will take up AB 3211.

Image manipulation and video doctoring have existed for nearly as long as photography and recording equipment, but they've historically required great effort and talent. Generative AI, or Gen AI, has changed that. Our ability to understand the fundamentals of our society and ensure basic safety. Undressing apps are rampant. These applications allow a user to create non-consequential, intimate images, also known as deepfake pornography. Creating convincing deepfakes takes less than 20 minutes and costs only a dollar.

Women and girls in particular, are vulnerable to this harm, and it has already been seen in California's high schools and middle schools. Technology is evolving rapidly, and according to investigative work of the New York Times, the models are updated so frequently that a chatbot that struggles with a task one day might mysteriously excel at it the next. Google has been public that their synth ID watermarking technology is in beta testing.

Academics are making strides in metadata and stenography, and companies have made commitments to this Administration at the White House and international standards-making bodies. But we, as policymakers, need to inform the pace of this development and the expectations for public disclosure. This ill is a broad framework to legally require companies to take on the responsibility of protecting our information environment, our children, and our democracy. The issue of generative AI-produced content can be tackled through a three step plan that was outlined in the committee analysis.

First, all artificial content must be labeled as such. Second, all real content, such as video and audio recordings, must be labeled as such. Third, the large online platforms must be required to prominently display these labels, allowing users to distinguish between artificial and real content. As the committee analysis stated, the bill seeks to enact the plan described above. It would one require the providers of GenAI systems to enable watermarking, detailing content provenance into all Gen AI outputs.

Secondly, digital cameras and other recording devices to offer users the option to embed water markings into outputs. And lastly, social media companies to prominently display the labels that contain the provenance information of real and Gen AI-produced content. I believe I have folks testifying. They should be coming. I do have two people testifying. Maybe we could move on to opposition.

Sure.

I know we went a little bit out of order, so. Ken Wang with Legal Counsel as cited, will be here, who's the sponsor of the bill, and David Evan Harris, formerly of Meta and currently of the chancellor's public school. Sorry, Chancellor's Public Scholars at UC Berkeley will be here as well. But we can move on to yeah.

So we can go to opposition. We will come back and give them their time when they're here. So opposition, please join us. We'll go a little out of order since there's two. We'll do two minutes each for both support and opposition when you're ready.

Good afternoon, Madam Chair, members of the committee. My name is Khara Boender, testifying on behalf of the Computer and Communications Industry Association and respectful opposition to AB 3211. CCIA is an international not for profit trade association with about two dozen members from a range of communications and technology firms. While we understand lawmakers interest in mitigating potential harms associated with AI-generated synthetic content. As written, the bill raises several concerns.

We are still working through reviewing the recent proposed committee amendments and look forward to providing additional feedback as needed. The bill's prescriptive requirements, particularly because of the current technical limitations for watermarking and digital provenance technologies, is problematic. Many of our members are working to implement tools to better detect and label AI-generated content, but the tools that are currently available are not always reliable or accurate, and while such technology is evolving, so are the means for bad actors to evade such detection.

Further, the bill raises questions about how to effectively watermark AI generated text or voice responses. For example, when asked a question and provided a written response, a large language model is unable to watermark AI-generated text, and commonly used voice assistants such as unlocked mobile phones and smart home devices produce AI-generated voice responses, but we cannot watermark those voice responses either. In the ever-changing AI landscape, we suggest that the bill be targeted to high risk cases.

For instance, responsible digital services are proactively implementing safeguards to swiftly remove illegal or dangerous content like AI-generated images depicting CSAM. By persistently addressing these well-defined and agreed upon risks, platforms foster an environment conducive to harm mitigation without impeding the innovation ecosystem. Limiting provenance and watermarking requirements to high-risk applications also helps guard against counterproductive unintended consequences. There are risks of notification fatigue or banner blindness.

This has been prevalent with website operators' attempts to comply with the EU's website cookie disclosure requirements, which showed that overly broad disclosure disclosure obligations could be highly disruptive while providing limited consumer safety benefits. On behalf of CCIA, I appreciate your consideration of these comments, and thank you for the time.

Thank you. We'll finish opposition part. Okay, perfect.

Okay, thank you Madam Chair and members of the committee, my name is Jason Schmelzer with Shaw Yoder Antwih Schmelzer & Lange, and today I will be doing my best Dylan Hoffman impression on behalf of Tech Net. We are respectfully opposed to AB 3211. While we agree that the intent of AB 3211 to create greater trust is a good one, we're concerned about many aspects of the bill. We're still reviewing the amendments, but believe there are still a number of issues that need to be resolved.

But we look forward to working with the author and potentially providing amendments that help us out in this situation. However, as the bill is in print presents multiple problems that require platforms to comply with the technologically infeasible and impossible standards. First, while we understand that the desire to regulate first, while we understand the desire to regulate an emerging technology, this is an area that would benefit from federal standards and regulation rather than a state by state approach.

Many of our companies and platforms are at the forefront of developing content provenance and watermarking technology, which is still in its early stages. As a general note, content provenance in watermarking is incredibly unreliable and in many cases, easy to break. Researchers at the University of Maryland were able to break all of the currently available watermarking methods. Some can be avoided by simple cropping, resizing, or screenshotting of an image. More concerning, the researchers were able to insert fake watermarks and credentials into images, creating false positives.

Any bill on this topic should account for this unreliability, especially when considering the overly punitive fine structure within the bill. If laws exceed the industry's technological capabilities, consumers might place too much trust and credence in watermarks, making them more susceptible to bad actors attempts to deceive. As it relates to this bill, AB 3211 enacts requirements for technology that is still under development and is rapidly evolving.

For example, there isn't a system that can break a system that can break a watermark text, making the bill's requirements to do so impossible to comply with. Furthermore, in its standards for large online platforms, AB 3211 should more clearly delineate between first and third-party content. First-party content would be images, videos or audio that is generated using a covered provider's generative AI tools and then is posted for distribution.

Dylan would want me to tell you you're over your time.

Thank you. For these reasons, we're respectfully opposed at this time, but look forward to working with the author in coming days.

Thank you. Thank you. Okay, we'll move on to support, and then I'll call both support and opposition to the mic. We're just gonna do this all out of order.

Good afternoon, chair and members, and apologies for our tardiness. My name is Ken Wang and I'm representing the California Initiative for Technology and Democracy, or CITED for short. We're proud to sponsor AB 3211, which would require watermarking and provenance data for content generated by generative AI systems. CITED which is the newest project of California Common Cause came together to seek state level solutions to the threats that disinformation, artificial intelligence, deepfakes, and other emerging technologies pose to our democracy and our elections.

We approach these issues through a cross disciplinary approach advised by leaders from civil rights and civic engagement, law and public policy, industry and tech, and more. To that end, cited is proud to sponsor AB 3211, which would add an important tool to our arsenal to help stem the tide of AI-generated disinformation. The committee analysis details several disturbing examples for how generative AI is currently being used to create non-consensual intimate imagery, perpetuate financial scams, and swing close elections.

Given the scale of the challenge before us, various efforts have already been initiated to address the issue. For example, the EU's new AI act is leading the way by requiring companies to develop content authentication and provenance mechanisms to tackle AI-generated disinformation, but domestic work efforts have been far weaker. Federal action has been limited to the Biden administration's Executive order on AI, which tasked the Commerce Department to develop guidance for content authentication, and various tech firms have also committed to voluntary efforts to establish industry standards.

CITED believes that California has the opportunity, as the home of Silicon Valley and the heartland of AI innovation, to lead the US's response to these threats. AB 3211 represents the important first step for our state to shape the solution. So lastly, I'd like to thank Assemblymember Wicks for your leadership on this important issue and for the chair and the committee staff for your hard work on the amendments. With that, I'll turn it over to CITED's technical advisor.

Chair, and members of the committee. It's an honor to be able to speak with you today and to convey my strong support of AB 3211. The California provenance, authenticity, and watermarking standards my name is David Harris and I'm here in my capacity as a senior policy advisor to CITED. I am also a chancellor's public scholar at UC Berkeley, a senior advisor for AI and elections to the Brennan Center for Justice at NYU.

I have advised the White House, the European Union, and the United States Congress about AI policy, and I also currently sit on a task force at NATO on AI and disinformation. AI is a powerful tool that can bring benefits to so many people, but it also, if not governed judiciously, has the power to cause great harm.

I previously worked at Facebook and Meta as a research manager on the civic integrity and responsible AI teams, where I carefully studied so many of the harms that AI can bring. You've already heard about the harms of AI generated, non consensual intimate imagery and child sexual abuse material that can be produced by AI. Misinformation campaigns can also be supercharged by AI. We're familiar with these campaigns from the 2016 efforts by Russia to interfere in our elections through social media.

The Department of Homeland Security has already warned us that China, Russia, and Iran are likely to use generative AI to target our elections. Watermarking is the best available tool we have to allow us to identify what content is produced by AI and what content is authentic. AB 3211 requires that AI companies make good on the promises that they have already made to add watermarking to their AI systems.

These promises, promises made by many of the members of TechNet and net choice were made first at the White House. In the White House, voluntary AI commitments from last July and were made again in public at the Munich Security Conference's AI elections accord, again signed by many of the members of TechNet and Net Choice, where a commitment was made this February to implement watermarking features in AI systems.

Given that it's already been nine months since the White House AI voluntary commitments were made and we've seen little progress, I believe that we have no other option than to legally require companies to take on this responsibility and protect our information environment, our children, and our democracy.

Thank you so much. Now, anyone here in support that wants to come up to the mic and weigh in to support first?

Trent Lang California Clean Money Campaign and strong support.

And for support and opposition, just name organization and position, please. Thank you.

Daniel Pearl on behalf of the American Federation of State, County Municipal Employees in strong support.

When to do it is always a question. Tracy Rosenberg Oakland Privacy we are support if amended, and we are checking out the committee's commitments.

Thank you. But you generally love our amendments.

Shane Gusman, on behalf of SAG-AFTRA in support.

Thank you. Okay, now anyone in opposition, please join us. Name, organization and position.

Yeah. Cameron Demetre with Capital Advocacy on behalf of Net Choice, we like to align our comments with CCIA and Tech Net. Thank you.

Hi. Ronak Daylami with CalChamber. Also align our comments with the opposition. Thank you.

Thank you. Okay. Seeing no additional witnesses here in the room, I'll bring it back to the committee. Before I do, I want to take a moment of personal privilege and just commend committee staff on the excellent work on all of the bills up today, the analyses, the amendments, all of it, I think was really phenomenal. So with that, Miss Irwin.

Thank you very much. And assemblymember Wicks, I know you have done, you've taken on all the big tasks in protecting our children on the Internet, and it is really critical, especially the previous presentation was on mental health, and we know how they are linked. And I do think that when you talk about deep fakes or AI generated images, there's a threat to democracy and a threat to our elections and, and just that Americans don't.

I mean, a threat to our children, too. But what, what we're hearing, I mean, I absolutely agree that everything should have a watermark, but what we hear from the opposition is that it's very easy to fake watermarks. And it's, you know, that the technology is not quite quite there yet. And so I agree with everything you're saying, but how do we get Americans to trust watermarks if it's so easy to overcome a watermark?

I would actually, if the chair's okay. Actually, my witnesses to answer because of their technical expertise.

Yeah, please.

David, thank you so much. That's an excellent question. And that's one of the reasons why the approach of AB 3030 to 11 requires watermarks to be inserted not only in synthetic content created by generative AI systems, but also in content that is authentic, that is created by the camera in the Ipad sitting right in front of you, or by your phone, or by any recording device.

It's very important that we attack the problem of AI-generated content that could be used in abusive ways from both sides, both ensuring that we can identify clearly and easily what is real and what is fake. There is no currently perfect solution for AI watermarking that makes it absolutely impossible to circumvent.

But this is a case where we should not let the perfect be the enemy of the good, and we should be very careful in listening to people who say, it's not good enough, so we should do nothing. And the same sense, if we ever found a seatbelt that could prevent every death in every automobile accident, that would be amazing. But that doesn't exist. Does that mean we should eliminate laws requiring seatbelts in cars? I don't think so.

Just because a solution is not a silver bullet to every single problem associated with AI, does not mean that we should throw up our hands and give up entirely.

No, I don't think we should give up entirely. That's certainly not what I'm saying. And I know that this Bill will continue to move along, but the question is, if initially you have, I don't know, 50% trust that Americans can have, or Californians, 50% trust that something is authentic, that is problematic for later on, they might never believe anything that they see. If, if that's kind of their first introduction to watermarks.

Yeah, that's absolutely a valid concern. The good news is that there are a number of companies that have already developed viable solutions that are marketing and talking about them publicly. One of those solutions is from Google and it's called Synth ID. Google is a member of an association, the Coalition for Content Provenance and Authenticity, that has many of the major companies also that are members of, the groups represented here today. And Google has committed to leading that initiative towards deploying watermarks.

The problem is they and many of these other companies have been now committing to do this for nine plus months. And we have to ask, how long do we wait? There's also a problem with waiting to require it that, that actually benefits the companies that are not taking initiative punishes the companies like Google, like Adobe, like Microsoft, that are moving forward through this coalition for content provenance and authenticity, taking initiative, putting out possible solutions.

But they're getting punished because when the others don't go along, it doesn't work. And that's why the CEOs of these companies are calling for regulation. They are begging for regulation.

So if you say Google is at the forefront, how. How much can we trust their watermarks?

Yeah, the unfortunate thing is that we don't have a law in place yet that requires a company that deploys watermarks to allow researchers to study them and scrutinize them. That is one of the items in AB 3211 is that it requires companies that do implement watermarks to submit them to red teaming, which is an adversarial method of testing where people from outside try to break their watermarks and see if they can break them. And then they improve the technology.

That's why we've really designed this bill to be something that makes these technologies improve over time and gives them space to flourish.

And I 100% agree that we need to be able to authenticate content. I'm just concerned right now that the technology is, is not there. And so, but obviously we need companies to take responsibility and develop as quickly as possible. So. Yeah, I get that. I guess the one thing, go ahead.

I was just going to say. No, I appreciate that. I mean, we're trying to, I think, in a certain way, create a regulatory framework and get ahead of some of the concerns that we have, right. That we're already seeing.

I mean, you're reading about it right now in the paper of, as mentioned, in middle schools and high schools, the non consensual pornographic imagery that's being created very readily and quickly and easily. So some of this stuff is happening obviously in real time. We also have a long legislative process. Right. And there's time. And I genuinely think from my perspective, having the input of the company is absolutely critical, which is why I shared language with many of them before the bill was even introduced.

I want them to be thought partners. I want to work with opposition to make sure we can do this in a way that's implementable. And if that means looking at implementation dates or other things that give us sort of runway to help kind of think this stuff through, I'm very open to putting all those things on the table. And the last thing I'll say, Miss Irwin, is you're one of the like thought leaders in the tech space. I'd welcome you in that, given your expertise. If you want to help me with that and work on that together.

No, I appreciate that. And I know you always work with all stakeholders, so this is definitely not criticism of that. It's just kind of this weighing of how do we get get there. So thank you.

Thank you, Mister Ward, thank you.

Thank you for introducing this bill. I know it's something we've had conversations prior to this year as well, how deeply you've been studying how to get ahead of the negative consequences of what could happen. And I think we share all of those goals as well too.

And I guess aligning with, but maybe also trying to simplify from my mind, some of my colleagues comments, well, so the watermark technology, and while it's not wholly worked out yet, but yet we're trying to define it and regulate it right now. What in your words, is a watermark?

Sure. I mean, from my perspective, it's a system that can embed invisible information to ensure you can identify what you're actually looking at. And there's multiple methods of water-marking and digital media, both, both the synthetic and the authentic. And it's not a new technology.

Right.

You know, many watermarking techniques used today are similar to those originally developed for the digital rights management. The system implemented 10 to 20 years ago to stop online piracy of music and other things of that nature. It's just more. It obviously needs to be more advanced now, but it's essentially the way I think about it in its most simplest terms, is creating like breadcrumb trails throughout the process. So we know when we can see if images have been manipulated.

So you know who's who, the origin of it, and you're able to sort of. Okay. And no, if we could ask. There's audio that's looping back here, so we're getting kind of an echo. There is the TV outside.

The TV outside is very loud.

I know. I feel like I'm in concert.

Yes. Although it's funny, I think you're actually louder in the hallway than in the room.

I know. No one wants me in concert.

And so we're still improving upon what this is and how to define it and therefore how we should responsibly address or regulate. But what. Talk to me about the European Union story. Right, because we're saying that this isn't already happening yet. I guess back to opposition's points, that we're putting the cart before the horse, I guess.

But we seem to be looking towards some success stories that are modeled on, well that are found on, I think a lot of deep thought that a lot of those policymakers have been able to put into it. And I will be the first to say on this dyess, that, you know, the reason why California is leading on this is because we don't have confidence that we're going to get federal legislation through and we need to be able to set that bar. That is why I think it's so important that she's. That we're working on this. But talk about the EU's maybe.

And Mister Harris actually advised the EU process. And so I think has a pretty intimate of how those regulations were put together. And I would with the chair's permission, defer to him.

Yes, please.

Thank you. These are wonderful questions and I want to give some context. I've visited Brussels twice in the last eight months to speak with the people involved in the drafting and implementation of the EU's AI act. That act is in the very, very final stages of being approved. The act does require watermarking of AI-generated content.

The way the European Union works is that there is an implementation phase that will take place during the one year following the final approval by the European Union's Council of Ministers, which is expected to happen in the coming weeks.

During that one year period, there will be a process that is referred to as co-regulation, where groups of companies will actually be convened alongside a European standards body called CEN-CENELEC that will eventually own a standard that is used to create the presumption of compliance with the EU AI act. And we've actually been, as the sponsor of this bill CITED, has been in communication with that team and had conversations to really understand the direction they're going.

What would happen if this bill passes is that we would have two parallel processes of implementation that would be roughly alongside each other between California and the European Union, where up until the moment when they are required for full implementation. And I think there would be great opportunities to collaborate and learn from what they've already learned by passing the requirement ahead of us. But in some ways, California would have the ability to jump ahead by passing this bill, depending on the final timeline passing.

Jumping ahead, but also sort of meeting European counterparts as well, which I would also argue for content that's created for information that is shared on a world wide web that, you know, we could probably be on parity to protect our, our users here in this state.

Yeah, and with the chair's permission. You got the gavel, Joe? Great. Mister Patterson's got the gavel. I think the other thing that's really important is that we do have conformity. I think, you know, when we have various states considering different things, Federal Government, European Union, there's a lot out there. And as much conformity as we can give around definitions and standards, I think it's better for the consumer, I think it's better for the companies, I think it's better for implementation.

And so that is really important to me as we work through this process and making sure that we're working and mindful of all the other regulatory entities that are out there.

And on one hand, like I'm, I'm sympathetic to the concerns that we don't have that definition yet, or we're still working on things. So it's very difficult to get ahead of that and predict that. But it's also happening over here in this space. And if we're just trying to conform to that, is this something that hopefully, as the year moves forward that definitions and application can be worked out. Do you have concerns with where this is going global, globally?

I'm happy to chime in a little bit on this, and I have to say as the fact that I work in the state capacity, I do have counterparts at CCIA who are based in Brussels and have been working with the European Commission there. So I'd be happy to follow up with more of the specific concerns that they have with regard to what's going on across the pond, so to say.

But I do know that in speaking with some of my other counterparts that, you know, there is some sentiment that because to the other witnesses point that this is not something that's already being implemented, it's kind of too soon to say what the other broader impacts of this are going to be yet. So by potentially going down that same route, there's still a lot of unknowns, especially given the technical feasibility limitations.

Fair. Thank you. The last observation that I want to thread into all of this is that are we trying to create a law that is going after, you know, the image itself, the actual content, or are we going over the. Trying to do something to prevent a negative action. Right. Is it the thing or is it the action? And for me it's about more.

We want to be able to make sure that there's no, there's no actions that are out there that are harming or going against any of our societal values or as to the moniker that the deep faking. Right. And so that's something I'm gonna be looking forward as this bill does kind of move forward because maybe 99% of the content that's out there for entertainment or for other purposes is completely fine and is obvious. Right. But it's.

And that action also could take the form of somebody that's getting around the technology of watermarking or having a fake watermark attached to it. And so what do we, what do we do about that when they meet the, meet the test of your bill by applying a watermark, but are still creating the harm.

Right. So there's a lot to think through on this, which is super interesting and I want to see it continue. But thank you for helping to tease out some of these issues.

Yeah, I appreciate that. It's, you know, it's a newer technology. You know, I think the manipulation that exists within these types of imaging really corrode our trust in the system. I think trust is really important. I actually see a lot of exciting value in AI and I'm genuinely excited that it's here. And I think it's going to mean a lot of economic growth and solving problems we haven't even figured out yet. But we have to have trust with that process, too.

And I think that's why we're trying to figure out a sort of surgical regulation here on a very specific piece of AI that's narrowly tailored. Because when you do have that manipulated content, I mean, ask the young girls who've had these non consensual imagery shared amongst all their classmates or elected officials who have deepfakes saying things that they don't believe in. I think you're right.

99% of the content out there is probably fine, but there's that 1% that that's what we're trying to really guard against here. And again, I genuinely welcome opposition on this because I'm not, you know, I'm not a tech, I'm a politician, right. I need the technical expertise from the companies, from the in house folks at the companies to help us think through how do you actually create a bill that can be implemented in an effective way.

Thank you, Mister Vice Chair. And then we'll go to miss Dixon.

Great, thank you. Just a comment. In terms of the type of deep fakes out there or, you know, content that could be created, I'm, you know, much more concerned right now about it being used for sexual content of minors or just, you know, sextortion or something like that. Obviously, democracy is very important, but oftentimes people can kind of figure that out at some point. But I wanted to make a, a comment and then ask a question to the author.

But the comments more to the witness you mentioned that, you know, if it's, you know, just because it's not a perfect solution, you know, doesn't mean we shouldn't move forward.

But I think the concern here is that, is that the false sense of security, you know, if I buy a car and I, you know, comes with seat belts, and they tell me the seat belts are going to, you know, close when there's an accident, whether or not that actually saves my life or not, you know, there's a level of assurances and a lot of regulation around knowing that that is going to happen.

And I'm not, I don't know if I have entire level of confidence that if there's an image out there that other technology, you know, foreign actors or something like that aren't going to be able to create some kind of watermark that makes it also seem like it's a, you know, safe product. And so I don't know if you want to respond to that, but I'm very concerned about the false sense of security.

May I have permission please? Yeah. Thank you. I think that's a very appropriate question to be asking. And I think that as this bill, were it to pass and become law, is implemented, you'll see different phases. And one of the phases is the phase in. Before people have updated their phones. Now software updates on phones are actually a very interesting part of this. I assume you sometimes download and install updates on your phone.

It's automatic.

It's automatic, perfect. That's because computer security is not a solvable problem. This is something that's discussed a lot in the security community. It's not something that you just finish, you never do. And that's why we have to keep installing these updates and keep improving the software. And that's why the language of this bill is intended in many ways to be what we refer to as self-updating. We use terms like state of the art to require that the companies use the best available standards.

And we also have built this, excuse me, this bill to facilitate and require the type of adversarial testing that mimics the state actors that you're discussing, that looks for get the smartest people in the room together, try to break the watermarks and try to make them as robust and indelible and difficult to remove as possible. And the problem is that we are not getting over the hump to doing anything to implementing this without the shove. We need the shove to get the companies to do this now.

They promised last summer to year do it. They promised this spring and it needs to happen. And without getting them over the hump, we don't have that opportunity. I spoke today with a company that's in the watermarking business that said if the companies were ready, they could deploy this in a matter of months. It could be deployed as a software update to your phone. So it's not impossible. It should be out there, it should be robustly tested, people should be attempting to break it.

And once we get into that phase where by default all real pictures you take, all real videos you take, all audio recordings you take, also have that signature of authenticity, then we get to a much better place where we have marking of authentic and synthetic content. And the lack of a watermark that proves the provenance of a piece of content itself becomes a symbol that that is not trustworthy. So a bad actor taking away a watermark is not as much of a thing threat.

Yeah. Okay, did you guys have a thought on that, or.

You know, this is, it's been mentioned a few times when we need to shove or it's sort of been asserted that the perfect is the enemy of the good or that we'd rather do nothing. And I just want to correct that notion. The witness himself has observed several things. One, that several tech CEO's have made commitments around this area. That these same tech CEO's have created an association to develop this technology and deploy it.

I do want to point out none of that technology has been deployed from my understanding. But we're also coming into committee today saying that we're working on amendments. So I do want to just correct the notion that a shove is needed to get us there or some other thing. But the penalty here is 5% of global revenues. So the discomfort is rational that we've sort of established that it's a nascent technology, that it's imperfect.

And while there is a very complex scheme here that's been developed, again, noncompliance for our Members is 5% of global revenue. So that leaves us obviously very uncomfortable. The last thing is NCII and CCM have been brought up. I would just point out that there are other bills that are very specific to those issues that are in the legislature right now. The Tech Net for one is supporting so.

Well, thanks. You actually kind of went into what I was next going to get to sometimes is being in the minority party. You kind of read the tea leaves that bills are going to get out of committee and so you want to see if there are ways that you can maybe work with the author to get it out despite the fact that bet you gets out.

I always welcome working.

Yes, I really appreciate that about you is that you do always check in with me to see where I'm at on bills and you know, it means a lot because you don't necessarily need my vote, you know, and. But, so I wanted to ask. I am very concerned about the penalty and if that's something you might be willing to work on. I could get to a support position today.

Yeah, I'm willing to work on that with you.

Okay.

And I also want to acknowledge the comments here. I mean, I do think I've been impressed with the not response because that's the wrong word. The proactive nature of the tech CEO's in this space. And a lot of what we're looking at is modeled off of their commitments. So I genuinely welcome them again in this conversation and welcome the companies in the conversation for sure, because we can't do it without them. Right. So I appreciate your comments and would love to work with you Mister Patterson.

Okay. I remember I used to be a staffer a while ago and the penalties and bills used to be like $10,000 per violation. Now I've seen $1.0 million. So you know, inflation's. We do have inflation, but it's the cost of living. Yeah, exactly. So I love to work with prices as well. With that commitment, I'll support this bill today. And I appreciate that.

Thank you very much chair. This has been an excellent conversation because we're dealing with something that we don't know how it's going to be fully formed. A couple questions. Have you, you talk to the tech people all the time, the tech executives and the key people. Have you had a sense you could come to a statement of principles as we go forward with Providence and watermarking? Is everyone aligned on this and approaching it similarly or in different ways? I mean, certainly they're mindful of what's going on in the EU. So are we coming to a standard. A uniform standard, an industry standard so to speak?

Yeah. Thank you Miss Dixon for the question. So most of the major companies in the space have joined something called the coalition for content Provenance and authenticity. Adobe is one of the leaders of that, but it's so many companies. Microsoft, Intel, Google joined more recently. Meta is not a Member, but does support the standard that they are putting out. And also the camera manufacturers, notably Sony, Nikon, Leica, Canon, have also joined this association.

The association currently has only deployed something that does not meet the standard defined in this bill because it's a marking, a digital marking, that is so easy to remove that you can take it off within seconds simply by taking a screenshot of that marking. But what my understanding from speaking to people who work at the companies that are members of this coalition is that the coalition has active working groups that are also developing standards that would meet the definitions of this bill.

And just to move back, to correct something from one of our other witnesses, Google has deployed a product called Synth ID that comes from Google DeepMind, their AI branch in the UK. And Synth ID, they clearly state on their website, is a tool for watermarking of AI-generated images and audio that even if you manipulate the image or the audio you take a picture of the picture, you re-record the audio that watermark persists, is robust and is to the greatest extent possible, indelible.

And the good news is that actually at the Munich Security Conference in February, when the AI elections accord was signed by 20 of these major companies, most of whom I've mentioned already, they specifically called out C2PA and synth ID as examples of the types of watermarks and the direction that they're going with the technology. So the associations are there. The problem is there's a first mover disadvantage here. Google's invested clearly a lot in this. Adobe has as well.

But then there are companies that are stragglers, notably Bytedance, last I checked, was not a member the owner of TikTok. We don't see Samsung participating in a meaningful way, and also we don't see Apple, which is surprising. But I do believe that if a bill like this re enacted, there would be in all likelihood a coalescing around the C2PA group and the standards that they have put forth and are developing, and they would advance much more quickly.

Okay, well, I hope you're correct. I mean, as I'm listening to you explain that all these different companies, I'm old enough to remember when it was DOS versus Apple versus IBM, and it was confusing for, I mean, just from a capital investment standpoint, and how do you. Whom do you trust and which is the better technology? It would seem to me this is so fundamentally important. I have a question also similarly with my colleague about the violation fine.

I think that's to be determined, but it would be magnificent. I mean, I know, Selma, women, you like California to lead the way. Okay. California will lead the way in whatever solution we craft here in this legislative process. It would be really not just fantastic, but it will be really important to 90% of the world's population who uses technology that we all can have confidence that all the players are aligned with a common platform, technology dealing with provenance, or watermark.

Apple may come out and call it something different, and it just would be. It would be more than helpful. It's almost essential that the world aligned the technology companies, the leaders in the technology, not adobe here and not somebody over here, because this is so fundamental. And I applaud you for doing this and creating this standard. I just wonder also, could we get those companies, every single Tech Net member, whomever, to agree to a statement of principles of what we're trying to align forward?

So when this comes to the floor of the legislature, both houses, you can say you have brought the industry together on something so fundamentally important that will protect our children and all users so we can have confidence in the system.

I think it's a great idea. I'm definitely happy to try. Miss Dixon, as I mentioned, you know, I shared the language with a couple of the major companies before it was public. I really wanted their input and their take and have done, have had numerous conversations, but with the trade association associations and with the companies themselves.

And, you know, I don't want to reveal any of those conversations because some of those are not public, but have gotten some good feedback from the companies, too, you know, and I think they are genuinely trying. I do actually believe that. But it's also ensuring that we have conformity across the system for them, I think is also really important.

Well, I agree, and I commend you for doing this. And it's a lot of, it's hurting rabbits here, but it's really essential. So thank you. Thank you.

Okay. Seeing no other members wanting to ask questions or provide feedback, I want to thank the author for her work on this. I think this was a very popular topic in the legislature. It continues to be, and I think it's a really important one. At the same time, what's been said here is true. Right.

This is a very young technology. One of the amendments here today sort of changed away from what Miss Dixon was saying to allow for changes in technology in the future, too, which I think is one of the important things as we make policy is none of this is going to be static. And so how do we create policy that allows for innovation and change that will hopefully make it more trustworthy and stronger protections for consumers?

But I think the author has mentioned many times today non consensual pornographic photographs. We've talked about the effects on our democracy in the recent elections throughout the country. We saw this kind of material propagating, and it is so critically important to the safety and security of this country and frankly, this world that people know what is fact and what is fiction. And I think that is what you are trying to achieve here today.

And so I want to thank the opposition for being a partner in moving us to that State of affairs. And I think that one of the things I really like about this bill, because I think it hits on a lot of what was said here today, is that it both requires the watermarking technology, but it puts some of the onus on the distribution channels, which is so, so critical, because that is where each of us will access this information.

And so making sure that they are a partner in ensuring that what we access is, we're aware of what's real and what's not. And so I want to commend you. I want to thank Committee staff for their extensive work on this Bill. I know they did a lot to get it to where it is today. And I know it will continue to change. And so with that, I respectfully ask.

I respectfully ask for an aye vote.

Okay, did we have a motion? No, we need a motion. Do we have a motion on the bill? So move. Okay, motion on a second. Let's call the roll.

Item number nine AB 3211 by assemblymember Wicks. The motion is do pass as amended to the Judiciary Committee. [Roll Call]

The bill is 5-0, so it will go on hold while we grab some members. Go and call. I see Miss Pellerin is next. Although Mister Albany said eat her here. So we will take up AB 2050. By Miss Pellerin, when you are ready. Welcome.

Thank you, Madam Chair and Members. In a state where everyone is mailed a ballot, the accuracy of our voter records is of paramount importance. The Electronic Registration Information center, known as ERIC, is a powerful tool elections officials can use to ensure that they have the most recent data possible to maintain the accuracy of their roles.

ERIC is a nonpartisan nonprofit governed by its Member states, which mutually agree to share data such as voter registration and vehicle licensee data, in order to maintain the accuracy of their voter rolls and identify eligible but unregistered citizens. When a voter moves, they often re register to vote in their new jurisdiction, but do not notify the prior county elections official that they have moved.

Due to the highly mobile nature of California's population, it is likely that there are millions of voter registration records which are now out of date due to a recent move. AB 2050 authorizes the Secretary of State to join ERIC on behalf of California. ERIC would provide the state with data on voters who have moved in state, moved out of state, or died. ERIC would also help identify Californians who are eligible to vote but are unregistered.

This would allow for more targeted outreach to be conducted by the Secretary of State and our local elections officials. ERIC security practices require that all sensitive data elements be subjected to a one way hash before they are transmitted to ERIC, while my technical witness can answer any more specific questions, in essence, a one way hash takes an input of any length and converts it to a fixed length output.

This process is specifically designed to be one way. ERIC Members do not have access to other Members data or reports stored on ERIC's servers. State voter registration systems are never connected to ERIC. ERIC is designed to work within the framework of the National Voter Registration act, under which states cannot remove voters from the rolls just because ERIC reports they have moved.

Under the NVRA, no registrant can be removed based on a change of residence unless that registrant either confirms their move in writing or fails to respond to a notice and fails to vote in the period between the notice and the second federal General election.

As a result, if a California voter was flagged as having moved in advance of the 2018 elections, for example, their voter registration could not be canceled unless they did not respond to the notice, did not vote in the 2018 elections, or did not vote in the 2019 elections, and did not vote in the 2020 elections. However, they would be moved to the inactive file if they failed to respond to the notice. After they were on the inactive file for two federal General election cycles, the voter could be removed from the rolls.

Should the voter wish to vote at any time, they could avail themselves of California's same day or provisional voting options. Additionally, AB 2050 specifically prohibits the sending of AB 60 license data in order to ensure that no one who is ineligible to vote is included in the data set. As some of you know, Colorado mistakenly mailed registration notices to non citizens. This was due to a human error in the Administration of Colorado's data set by Colorado's elections officials.

Not a flaw in the ERIC system. And as you may all know, nine states have left ERIC in the last couple of years. I won't sugarcoat it. The states who have pulled out of ERIC have given in to conspiracy theorists, couch and data security concerns California's efforts to increase the accessibility of our voter registration system has been a great success. By the Secretary of State's estimates, we have almost 5 million eligible but unregistered voters.

Becoming an ERIC Member is a natural next step in California's mission to improve the accessibility and ease of voting. Furthermore, better data on which citizens are unregistered allows for more targeted outreach, maximizing what little funding we do have. There are some, including a former President, who oppose ERIC. With regards to some of the privacy concerns raised, we believe that California Consumer Privacy act already indicates a legislative differentiation between encrypted and unencrypted, dated under the CCPA, disclosures related to a security breach are not required if the personal information is encrypted.

The CCPA defines personal information as an individual's first name or first initial and last name when either the name or the data elements are not encrypted in combination with a Social Security number or driver's license number. As California would only be sending encrypted personal information to ERIC, so an encrypted Social Security number, date of birth, driver's license. We believe that under CCPA, this data can be shared in its hashed form.

AB 2050 is sponsored by the California Association of Clerks and Elections Officials, and with me to testify in support is Shane Hamlin, our Executive Director of ERIC, and Dave Mulligan, a cybersecurity expert.

Okay. Two minutes each, when you're ready.

Thank you. Good afternoon, chair Bauer-Kahan and Members of the Committee. My name is Shane Hamlin. I'm the Executive Director of ERIC or the Electronic Registration Information Center. I appreciate the opportunity to speak today in favor of AB 2050. ERIC is a 501 nonprofit. We are a Member led, Member funded organization. As you heard, our Members submit photo registration and motor vehicle Department data on a routine basis.

We are also certified to receive and use death data from the Social Security Administration, and we are a subscriber to national change of address from the United States Postal Service. Utilizing only these four data sets, we provide our Members with reports that identify out of date voter records, deceased voters, individuals who appear to be eligible but are not registered, as well as potential illegal votes. Members must use these reports in strict compliance with federal and state laws. ERIC has always taken data privacy and security seriously.

These principles are baked into our governing documents. We are generally aware of data privacy standards globally and here in the US, and we are watching as states adopt strong privacy protections modeled after California's excellent example, the ERIC Bylaws and membership agreement include extensive data privacy and protection provisions, and our technical infrastructure includes multiple layers of security to protect protect these data. Let me share a few specific examples.

The most sensitive data, as you heard, is hashed- driver's license, dates of birth, and Social Security numbers, and these data elements are essential for high quality, accurate list maintenance. They are hashed by our Members before they submit them to ERIC, and then, inside the secure environment, our staff hash them a second time. Additionally, Members encrypt the entire data file before transmitting it to Eric, and data at rest in the data center are encrypted. We follow internationally accepted cybersecurity standards.

We recently completed a type one SoC two audit, and we will do a SoC two type two audit later this year, and we have worked with state technology departments to provide assurances that our data handling practices are acceptable. Finally, I'll summarize four key provisions from the membership agreement. ERIC and our Members are bound by federal data protection laws that apply to the datasets we use. Members are prohibited from using the data or reports for anything other than the Administration of elections, with some lawful exceptions.

ERIC is prohibited from using data or reports for anything other than providing services to our Members, and any unauthorized use of the data or breach of the data must be disclosed to the Members or to ERIC, as applicable. I believe the current Bill draft addresses privacy concerns that were recently raised, and I also believe the letter you have from Professor Morris at the Carey School of Law at the University of Pennsylvania addresses very well recently expressed concerns about the use of ERIC data for list maintenance. With that, I'd be happy to answer any questions you may have. Thank you.

Thank you.

Good afternoon, Madam Chair, Members of the Committee. My name is Dave Mulligan. I currently work in the cybersecurity division of Fiserv, Inc. A major, excuse me, a major financial technology company. I've worked in this field for over 20 years, and in that time, I have helped dozens of companies to secure their most sensitive data and have analyzed many applications for security purposes. I should say, though, that today I'm here in my personal capacity, not speaking for Fiserv officially.

When I read the analysis prepared for this Committee, I was struck that in multiple places, no distinction was drawn between sending the plaintext value of a data element to ERIC and sending a hash of that value to ERIC. A hash function is a particular type of operation on data that by design is irreversible. The recipient of hashed data cannot reverse the process to obtain the original data, and there are no current known attacks on the SHA two family of hashing algorithms used by ERIC.

Using a hash in place of the value itself is a completely industry standard mechanism for comparing data values without actually needing the actual value. If, say, the same date of birth is hashed independently by two different states and sent to ERIC, ERIC can then compare those two hashed values, which will be identical, and know that the original values were the same without having any ability to know what that original value actually was.

The PCI DSS standards, which govern how companies have to handle credit card data, list one way hashing using strong cryptography as one of only three acceptable methods for protecting credit card numbers before storing them. And that's pretty much the gold standard. So, in summary, it's my professional opinion that the data handling procedures used by ERIC follow the industry best practices and don't cause any privacy issues, since sending a hash of a sensitive data element is not the same as sending that sensitive data element.

Thank you. Anyone else here in the room in support of this Bill? AB 2050.

Hello. Andy Liebenbaum on behalf of the County of Los Angeles and strong support. And I've also been asked to present the support of the California Association of Clerks and Election Officials who couldn't be here, but they also strongly support the Bill. Thank you.

Good afternoon, chair Members. Tanisha Boatner, on behalf of the California State Association of Counties, in support of AB 2050. Thank you.

Thank you. Anyone here in opposition of the Bill? Okay.

Ethan Naegler. On behalf of the City Clerks Association of, California, in support.

Okay.

My name is Colleen Britain. I'm with the election Integrity Project, California.

Did you. I'm sorry, just clarification. Did you submit a letter?

Yes, we did.

Okay. I just don't have it. That's why I stopped you.

Oh, you should have it.

Oh, no, I do have it. I do. I apologize. Okay, my bad. Please go ahead. Two minutes.

Okay. Accuracy accurately maintained voter rolls are essential to restore the faith in our electoral system and preserve our republic. ERIC is not the answer. It facilitates exploitation of vulnerable populations, especially non citizens. It prohibits the transmission of any data regarding voter eligibility based on citizen status. Yet ERIC maintains that states contact 95% of those on the eligible but unregistered list in an attempt to get them registered.

Unfortunately, the state has no way of knowing or determining which of those on the ERIC provided eligible list is actually eligible by law. As a result, ERIC could easily facilitate adding more ineligible individuals to the roles, the exact opposite of its stated purpose. ERIC puts the state at risk of violating three major federal laws.

First, the Help America Vote act, which says that state election officials shall implement a statewide voter registration list maintained and administered at the state level, and those same state officials shall perform list maintenance on the regular basis. There's no provision in that statute that authorizes any state to outsource these obligations to a third party. Second, the National Voter Registration act states that states shall maintain and make available for public inspection all records concerning activities conducted for the purpose of accurately maintaining voter lists.

In contrast, ERIC states that if a Member receives a request to disclose ERIC data and determines that it's legally obligated to comply, it shall not make that disclosure without first obtaining a court order compelling it to do so. Third, ERIC clearly violates the driver's Privacy Protection act, which states the fact that an applicant has declined to register will remain confidential. Last year's ERIC Bill died due to its cost. Cost is the reason enough to kill this unnecessary and unwise Bill again.

Lastly, there is no requirement in ERIC's bylaws that Member states must actually remove ineligible voters from their registration rolls. EIPCA strongly urges a no vote on AB 2050. Thank you.

Thank you. And I apologize for the mix up. Appreciate you being here and testifying. Yes, it is. I was looking at the wrong piece of paper.

It's a lot more detail than I could.

Yes, that's why we always make sure that people submit the letters cause it's helpful information. So, yes, no, I apologize for that. I was looking at the wrong thing, but I appreciate you being here testimony. Thank you. Anybody else in opposition to this Bill?

Come on up. You can give us your name. Hold on so you don't get hit by the door. So, name, organization or location and position. Thank you.

Hello, my name is Eunice Malley. I'm with the election integrity Project, California, and I strongly oppose AB 2050. Thank you.

Thank you. Come on up.

I can't do this. I'm Judy Friendo, I'm in Solano county and I'm also with the election integrity California, and I very strongly oppose and ask you to really consider this. Thank you very much. Thank you. Thanks for being here. Come on. Hello, my name is Nancy Stilts and I'm with Solano County. I work with the election Integrity Project California, and I strongly oppose AB 2050. Thank you. Thank you.

Good afternoon. Lee Mante Berry, coordinator for election Integrity Project California out of Placer County. I strongly oppose AB 2050. Thank you.

Thank you.

My name is Joel Berrio, I'm just a retired person, but I strongly oppose AB 2050. just a retired person. Thank you. My name is Rita Pizziali, I'm a resident of Placer County and I strongly oppose this Bill. Thank you. Careful of that door, yo. Yeah, yeah. Rochelle Connor. I am a registered voter in Solano County in opposition to 2050. Thank you.

NotThank you. Not seeing anybody else? I'll bring it back to the Committee. Any questions or comments? Yeah. Mister Bryan,

This Bill enjoys the co- Authorship of three Assembly elections chairs. I know you've been working on this since the day that you came into this body. Happy to make a motion to move this Bill.

Thank you. I'll second the motion. Thank you. Miss Dixon. You good? No, you good? Okay, awesome. Just want to make sure. So thank you. I also want to reiterate my gratitude for your leadership in this space.

I know you are one of the people in the body who has the biggest depth of experience in elections and ensuring that our elections are safe, open and fair to all. The integrity of that is so critically important to you. So I really appreciate you leading and helping us figure out the best path forward in that. And with that, would you like to close?

Well, I respectfully ask for your I vote. Maintaining accurate voter files is the utmost importance to maintaining our democracy.

And this is definitely an excellent tool for elections officials. So thank you. Great. Thank you, Miss Pellerin. Let's call the roll.

[Roll Call]

That has three votes, so we will leave it on call for absent Members. More people will come back. So thank you.

Thank you very much.

Mister Maienschein, AB 287. When you are ready.

How are you? Thank you very much, Madam Chair and Members, I want to begin by accepting the amendments as proposed in the analysis, California is facing a staggering number of overdose deaths. The impact of fentanyl alone has been felt in every one of our districts. AB 2871 creates a new tool to help combat this crisis and prevent future deaths. The Bill allows counties to create overdose fatality review teams composed of public health professionals, law enforcement, drug treatment providers, and local nonprofit organizations.

The review teams will examine the circumstances surrounding individual overdose deaths to identify commonalities or other actionable items that can lead to informed efforts to prevent future overdoses. The Bill allows members of the review team to share relevant information regarding a decedent within the scope of the review team. In order to protect privacy, the Bill makes all communications by or within the review team confidential.

The Bill is modeled off of existing review teams such as the child death review teams, domestic violence death review teams, and elder and dependent adult death review teams. With me to testify in support is Doctor Steven Campman, the San Diego County Chief Medical Examiner.

Thank you.

Good afternoon. Chair Bauer-Kahan, sorry, and Vice Chair Patterson and Members of the Committee. Again, I'm Doctor Steve Campman, the Chief Medical Examiner for San Diego County, and it's an honor to testify before you today, and San Diego County is proud to be the sponsor of this legislation that will give counties an additional tool to address the drug and opioid crisis locally. According to the data from the State's Health Department, opioid related deaths increased by 121% between 2019 and 2021.

The alarming trend is true across other counties in the state, including our county, San Diego County. In 2021, 1,309 people died in our county because of drug overdoses, 896 of those due to opioids, and among those over 800 due to fentanyl alone. AB 2871 would allow counties to create overdose fatality review teams to engage in system wide team review and information sharing across county agencies, local stakeholders and experts when there is a drug fatality with the sole purpose of preventing future opioid overdose deaths.

Currently, information about those that have died is largely siloed between different agencies, despite evidence that the other committees that were mentioned for children, domestic violence and elder abuse have had success. Overdose fatality review teams would improve systemic issues by streamlining the ability to share information about cases to help identify opportunities for improvement at both the system-wide and the individual levels. By allowing stakeholders that may have had contact with the deceased to work together to identify gaps, especially in delivery of service.

AB 2871 would allow for information sharing while still protecting the information shared for these purposes in accordance with confidentiality standards. To kind of better explain the information sharing in the context of this kind of committee, the members sign confidentiality forms and then the members choose a topic, perhaps, you know, school-aged kids or fentanyl, people that have died with fentanyl, to focus on for a meeting. And then those present bring the information that they have about that person to the meeting and verbally share.

So we're not. It wouldn't be duplicating people's medical records and sharing them about the Committee. The Bill would help us create a database of de-identified cases to assess health outcomes more comprehensively and strategically invest resources towards prevention. AB 2871 would ensure continued compliance with confidentiality standards, still allowing the overdose fatality teams to identify trends and systemic barriers to prevent overdoses in the future. I respectfully request your support of the Bill and thank you for your time and attention.

Thank you. Anyone else here in support of this Bill?

Good afternoon, Madam Chair and Members. I'm Elizabeth Espinosa, today expressing the support of the Urban Counties of California, as well as the County Health Executives Association of California. Thank you.

Thank you.

Clifton Wilson on behalf of the Fresno County Board of Supervisors with the position of support if amended. Thank you. And thank you for raising the mic.

Clifton, it was still too low for you, but we're trying. Anyone here in opposition to this measure?

Seeing none. I will bring it back to the Committee. Any questions or comments? Seeing none. I just want to thank the author. I know the most important job in your life is father. And that you are someone who really takes seriously protecting our communities. And this is a huge step in the right direction, because the overdose epidemic, especially amongst our young people, is something we need to throw everything we've got at. So I want to thank you for this.

And with that, would you like to close?

Well, thank you very much, Madam Chair. I would use that as my close. This is something that's going to be an ability for us as lawmakers, both at the state level and the local level, to craft better policies that prevent this from happening. So with that, I respect the request an aye vote.

Great. Do we have a motion on this measure? Great. Madam Secretary, please call the roll.

Item number four, AB 2871 by Assembly Member Maienschein. The motion is do pass, as amended, to the Appropriations Committee. [Roll call]

That Bill has five, so it's on call. We'll leave it open for absent Members.

Thank you.

Mister Alanis, you are up with AB 3080.

Thank you. Hi, guys. Well, thank you, Madam Chair and members; let me start off by saying that I, indeed, am accepting committee amendments today if need be. Thank you, Josh. And I'm grateful to your committee staff for the time and thoughtfulness they put in for making this bill better and narrower. As we begin the discussion on this very complex and sensitive topic, it has not lost on me the very real concerns that the opposition has.

I have read their letters, and I fully intend to continue to discuss and work through these very important issues. But what also weighs in on me is my times, the crimes against children's detective with the Santos County Sheriff's Office, and where I saw what I believed to be an unspoken and often hidden but scientifically supported impact on pornography. While I don't claim to have all the answers, and I'm certainly not an expert on age verification technology, that's what these guys are here for.

This bill, along with the amendments I'm accepting today, is measured. Well, let's just say that it's going to make sure then, and I hope everybody's happy with this bill. And with that, I'm going to bring up those that were testifying. Heidi Olson, pediatric nurse, and examiner specializing in sexual assault cases. Iain Corby, Executive Director for Age Verification Providers Association, and Joseph kohm, the Policy Director for the - I'm sorry. Need water. Policy Director for a Family Policy Alliance.Thank you.

Thank you. So we've given prior permission to do two minutes each, but just for clarity, the opposition will have the same. So when you're ready. All right.

Good afternoon, Chair Members. I'm Heidi Olson. I'm a pediatric sexual assault nurse examiner, and over the last few years, I have performed or reviewed over 1500 forensic exams of kids who have been sexually assaulted. A really disturbing trend that we're seeing at our hospital, and that's being seen around the United States, is that the age range most likely to commit sexual assaults right now are 11 to 15-year-old males.

Now, while problematic sexual behavior is a multifaceted issue, no doubt there's a clear connection between repeated exposure to pornography and child-on-child sexual assault. There are over 100 research studies showing there's a correlation and causational relationship between pornography and violence. Kids don't have fully developed brains. They're really susceptible to what they see in pornography. This is problematic because mainstream porn today is very violent and degrading, and it's the number one sex educator of our kids today.

A recent study estimated that two-thirds of adolescents are currently addicted to online porn. Kids are accessing violent porn at mind-blowing rates. I just talked to a rural high school that monitored their 140 students school-issued devices, and in one month, there were 13,000 hits to Pornhub. We are seeing numerous negative effects on tweens and teens, from erectile dysfunction to glorifying sexual violence to increased child sexual exploitation and the fundamental misunderstanding of consent, as pornography does not value intimacy or mutuality. The porn industry has a goal of making money. How do you do that? You create lifelong users. Research shows that the earlier a child is exposed to pornography, the more likely they will be lifelong users. So, for those reasons, I strongly support this bill. Thank you.

Thank you. When you're ready.

Thank you. Madam Chair and members, I'm Iain Corby. I'm the Executive Director of the Age Verification Providers Association. We represent 30 companies around the world that provide online age-verification technology, and our mission is to make the Internet age-aware, not identity-aware. So online age verification can be done, and is already being done at great scale. It can be done anonymously, effectively, inclusively, conveniently and cheaply.

So, if American technology can put a man on the moon, we can certainly allow you to prove your age online without disclosing your identity, and that's the essence of age verification. If we wanted to do identity, we'd leave that to the much bigger identity industry. So the basic way this is accomplished is by proving your age to an independent third party, and then that independent third party simply says yes or no to an adult site as to whether you're over 18.

We believe age verification providers should be independently audited against approved international standards, which not only check that they're getting the results right but also that personal data is being processed securely and is not retained after the check is completed, as this bill, as rewritten by the Committee, explicitly requires. There's an ever-growing number of ways to prove your age. You might use a piece of government documentation or a reusable digital identity, which we compare to a live image to make sure that it's you.

You might get data from your cell phone network or utility provider, or your bank, and recently, we have developed a way of estimating your age based on just what you look like within a year and a half of your actual age. And the Federal National Institute of Standards and Technology has just completed some, some very rigorous testing of that, which was, and they've published some promising results.

It takes just a few seconds, and we're cooperating as an industry to create interoperable, reusable age checks through a project funded by the European Commission, originally called EU Consent; this innovation is driving down the cost. The UK government estimates about 12 cents per check, and we agree with them that that price will only go down. So, thank you very much for the opportunity to speak to you today.

Thank you. We have this conversation a lot, and that was helpful to the last witness.

Good afternoon. Chair and members of the committee, my name is Joseph Kohm. I'm an attorney and the Policy Director for Family Policy Alliance, which is a nonprofit organization that focuses on advancing family values through legislation around the country, and we're proud to strongly support AB 3080. So, with the rise of technology, children have nearly unlimited access to adult websites containing age-inappropriate pornographic material. Research shows that by the age of 17, about 75%, three-quarters of adolescents have been exposed to pornography.

Research also proves that pornography harms children. Studies show that over 88% of pornographic videos depict some kind of sexual violence, while over 48% of these same videos include some kind of verbal abuse. Studies also show that 53% of male adolescents and 39% of female adolescents actually believe the lie that pornography is an accurate representation of sex. So this means that adolescents are learning about sex from a perspective which portrays sex as physical abuse.

Now, additionally, research also reveals that out of those individuals who have been exposed to pornography, 58% access the material unintentionally. So what this means is that adolescents are becoming addicted to pornography through accidental interaction with certain websites that age verification and parental involvement could have prevented. Now, I want you to know that California is not the first state to propose a bill like this. Last year, seven states of varying persuasions passed similar legislation, and another eight have passed them through their legislatures.

This year, most of these bills have been bipartisan efforts, including somewhere the legislative houses voted unanimously to pass the bill. And most of these bills have withstood judicial scrutiny. Most recently, the Fifth Circuit Court of Appeals upheld one of these bills. In a landmark decision, the court made clear that states have a compelling interest in protecting children from Internet pornography and that bills like AB 30, which is substantially similar to the bill at issue in that case, do not violate the First Amendment.

So right now, Chair and members, the average age of early of exposure, first exposure to pornography is between seven and 13. So AB 3080 is critical legislation necessary to protecting children and empowering parents. We strongly urge you to support, and I'm here for any questions. Thank you.

Thank you. That was an impressive, exact six minutes, team. So I'm gonna.

We've been timing it.

Kudos to the support on this bill. Anyone else in the room in support of this bill. Welcome to come up, register your support, name, affiliation, and position.

Rochelle Connor, on behalf of Concerned Women for America, and strong support.

Thank you.

Hi, my name is Mary Lentz. I'm an associate producer for Disney and Hulu. I'm strongly supporting this bill.

Thank you.

Hi, my name is Jewell Baraka. I'm a survivor who was trafficked in porn at 14, and the author of the recent book Coming of Age on a Porn Set, and I am strongly in support of this bill.

My name's Helen Taylor, Vice President for the anti-trafficking organization Exodus Cry, in full support of this bill. Thank you.

Thank you.

Jenna Mckaye, a human trafficking survivor, and with Exodus Cry, and I support this bill.

Thank you. Watch that door. It keeps opening behind you guys. But please join us.

Ashley Faison, I'm the founder of an anti-trafficking organization, Diamond Collective, and also a survivor in support of this bill.

Thank you.

Hello, I'm Meziah Maddox, and I'm a parent, and I support this bill.

Thank you.

Hello. My name is Peggy Delgado Fava. I'm the Executive Director for Bridge Network, and we work with marginalized, at-risk youth. Support of this bill. Thank you.

Pastor Brandon Campbell, Northern California Director for California Baptist for Biblical Values, and support.

Thank you.

Christine Campbell, just a concerned parent and support.

Thank you. Not just a concerned parent.

Greg Burt, Vice President of the California Family Council, in support. Thanks.

Thank you.

Brock Campbell, on behalf of Lighthouse Baptist Church in Santa Maria, in support.

Thank you.

I'm Emily Campbell from Santa Maria Lighthouse Baptist Church and I'm in support.

Thank you.

Hello, Dr. Doug Bennett, founding Executive Director on behalf of Magdalene Hope, Restoration Ranch Women's Shelter, Rescue Grounds Coffee Company, and people for the Abolishment of Trafficking Humans. I'm a former porn addict and sex buyer, and passing this Bill might help a young man go down the same road that I went down. So I urge you for an aye vote.

Thank you, sir. Do we have principal opposition in the room that wants to come up and testify? Oh, okay. Oh, there we go. No, we have someone. She's coming. Join us. Okay.

Just a quickie. Samantha Corbin, on behalf of the Electronic Frontier Foundation, in opposition to the measure. Thank you.

Thank you.

I'm happy.

Please, there's six minutes for the opposition. That's equal time around here. That's part of our democratic process.

Thank you, Chair Bauer-Kahan. My name is Alison Boden, and I'm personally from Sonoma county.

Can I stop you 1 second? Because I think Oakland Privacy might want part of the time. Do you want to share time? Okay, so if you guys want to share time, I just want to make sure we coordinate. Is that okay with you? Yeah. So three minutes each. Does that work? Sure. Perfect.

So, my name is Alison Boden. I am the Executive Director of the Free Speech Coalition, which represents people and organizations engaged in constitutionally protected sexual expression. We have hundreds of Members throughout California, places like Lafayette, Oakland, San Diego, Long Beach, and, of course, Los Angeles. I am here to oppose AB 3080 because it's severely flawed, both legally and technologically. Legally, the Supreme Court's already ruled that requiring age verification to access protected expression online is unconstitutional.

This is why my organization is challenging similar laws in other states and why it's important to note that the Fifth Circuit decision is being appealed to the Supreme Court as we speak. Technologically, the zone entertainment industry actually supports efforts to keep minors away from our content. But unfortunately, this Bill is ineffective and has an ulterior motive.

As I'm sure Members of this Committee are well aware, restrictions like this are easily evaded by VPN, which is a simple technology that's already used by nearly half of minors to change their location when they're browsing the Internet, like Netflix. But the fact is, AB 3080 has nothing to do with protecting children. It's about restricting the kind of content that Californians of all ages can access.

The language of this Bill vastly expands the definition of obscenity, flying in the face of Supreme Court precedent in Miller versus California. Instead of protecting children, this bill will subject all manner of people, businesses, organizations, libraries, to ruinous civil liability for allowing access to indecent material. If they can't afford to implement costly age verification technology and giving them no ability to defend themselves, should they be sued because they can't retain proof that they identified any particular person?

This bill has gone into effect, or one like it in other states. And what we've learned is that people don't comply. Upwards of 80% of people, when presented with the option to age verify, leave a website and go out and find one that doesn't comply with laws. This isn't the same as flashing your ID to buy beer. It's transmitting highly sensitive and valuable information over the Internet, exposing it to all types of danger.

The Supreme Court ruled in Reno v. Ashcroft that, no, it's Reno versus ACLU, that the government can't require age verification where less restrictive means of presenting miners from accessing inappropriate content existing. They did then, and they do now. This precedent has been affirmed again and again in courts all over the country. And just last week, Arizona's democratic Governor vetoed a similar bill because it violated the First Amendment.

We want minors off our sights just as much as you do, and we're happy to work with this Committee to come up with an effective constitutional solution. Thank you very much, Chairwoman and Members and staff.

Yeah.

Oh, hi again. Tracy Rosenberg from Oakland Privacy. I'll be brief. We just sort of want to highlight the fact that we don't need to guess what this would look like if we were going to go forward and adopt it, because it's already happened in seven states. And what has happened in those seven states, most of which are states, a lot of whose laws we don't actually unlike so much. Is that how to say this?

The most legitimate and safe pornography options for both adults and youth, have vacated the state they've left. Pornhub is gone. That means that anyone searching for porn in those states is essentially getting pushed to sites that are shadier, less safe, worse cyber security, much more snuff, much more non consent porn, much more dangerous stuff. I mean, I am not an advocate for Pornhub, God forbid, but they have made progress over the years in getting some of the worst of the worst off their site.

So what is the incentive in pushing people that are looking for pornography away from sites that are safer to sites that are far, far, far less safe? Are we really accomplishing what we want to accomplish here? And moreover, porn is not like a small, teeny, tiny segment of the Internet. We tend to kind of think that, but actually, 25% of all Internet searches are for porn. And that makes me sad because it's a really high number. Yeah. Like all the people in this room.

Like one out of four, you know, but those are the numbers. So it's a huge amount of Internet traffic and it's going to go somewhere. And so I think that we have to be realistic about. I mean, yes, we absolutely don't want kids to see porn. We don't want kids to see dangerous porn. Absolutely. But we have to be a little realistic about what we're choosing to do is act, you know, what the impact is going to be on the ground.

Thank you.

If we're encouraging people to do that, it's not good. Thank you.

Thank you. Anyone else here in opposition to the Bill?

Good afternoon. Cameron Demetre with capital advocacy on behalf of net choice. We have just two concerns real briefly, first amendment concerns with obscene and indecent the definitions. And to the case that was referenced in ACLU are Reno versus ACLU, and also data minimization concerns. Thank you.

Thank you. Seeing no other witnesses in the room. Bringing you back to the Committee. Mr. Bryan.

Learning so much today. Question, just for the author, what did you think of the opposition's points?

Well, I think just one, talking about 25% of the usage, I mean, makes me even want to do this even more if it's that high. I didn't know it was that high. If you mind, if you want any of my experts to speak on it as well. We could do that.

Just wondering if there's more to discuss with the opposition after if this Bill were to move through Committee hearing, if that would be something you'd be open to.

If I may note one thing, it sounds like there wasn't a chance, and I understand this, to review some of the amendments on the definitions around indecent materials that hopefully may address some of the concerns. So hopefully the opposition can review the amendments and see if it addresses some, maybe not all of their answers.

And there's still more to come. Thank you.

Thank you, sir.

So the Bill now doesn't talk about indecency at all. It uses the term sexually explicit content, which is much narrower. So hopefully that addresses some of what we heard today. And I want to thank Committee staff because they went a long way in working on the First Amendment concerns, and hopefully, and if it moves out of Committee today, it will go to judiciary, which obviously, First Amendment is a huge part of what they will be looking at. Miss Dixon.

Thank you. Chair. Technical question, the identification, the verification process, how would that work?

Thank you. Jumping around a little. That's a great. There are a number of different ways of doing it. The traditional way would be to use a government issued ID. The technology is such now that we can do that on your device so we don't actually have to transmit that information away from the phone. Likewise with the facial age estimation, that also can be processed entirely on device. And then we use what's known as privacy enhancing technologies, PETS.

So that when you are giving the confirmation to the adult website that the user is over 18, they see absolutely no information about that user. And as this Bill requires, after being edited by the Committee staff, we're not allowed to retain any of that data and would be subject to heavy fines if we did.

So. Thank you. So who receives that transmittal of the license or the identification? Who does, who gets it, what machine?

So I'm saying it is possible at the moment to do that locally on your own device, so it never leaves. Until recently, that has traditionally gone to an age verification provider. So we have an IEEE standard, a British standard, and there's an ISO, an international standard against which all our providers are audited. And part of that is confirming that they look after your data while they're doing that process, just as well as your healthcare provider or your bank.

That's true. Well, I think that's a vital part of this to make it. It's the first time I've ever heard any possible practical, technical solution in order to check identification, other than algorithms and everything else that change every year, where the age would change every year, whereas this could be a definitive privacy protected identification transmittal process.

If I might just add, the European data protection authorities have been intensely interested in our industry, and the French, Kinneil and the Spanish AEPD, have both produced very extensive new requirements, which we are conforming to now in order to meet the very high standards that they expect on privacy.

So you're saying that these are already. This is in place?

Yes. So we have a. Through the interoperable solution, which is a tokenized solution, which essentially means once you've proven your age, once you can reuse that same age, check in an entirely zero knowledge proof so that other websites will know that you're over 18 without you having to repeat the age verification process.

I see. Okay. Thank you very much.

Yeah, Mister Vice Chair.

Thanks. You gotta be careful what you Google when trying to figure stuff out on this. This is a government issued computer here. If an individual goes in. I missed most of the testimony, and I apologize for that, because I'm on four committees at the same exact time, so. But when you go into a grocery store and. Well, not you or, you know, maybe somebody, you know, purchases porn magazine, if that still happens. I don't know if that happens anymore. But back in the day, did you have to show ID like to buy?

Sometimes?

The answer is yes. And in fact, one of the cases that the Fifth Circuit relied on, in fact, the primary case was a case called Ginsburg. That's the short name for it. It was based on a very similar law, actually, from all the way back in the fifties that required the same thing. Basically, if you go into a bookstore that sells back then, what they called girly magazines or girly books, the all just had to flash their identification to let them know they were of age.

So what the court ruled is that what this bill and what the bill at issue in that case required is basically the online electronic version of the same thing, which is why it's not a substantial burden on adult speech. Adults can watch as much porn as they want to. Once one of these bills goes into effect, all this does is protects children from coming into contact with it.

I mean, if it's okay at the Chair, I'd love to hear from you.

So Ginsburg actually dealt with a magazine salesman who wanted to sell porn to minors. This is not what this Bill deals with and it is actually a very different thing to flasher ID to someone who isn't memorizing it as it would be to maybe pass it around to every single person that now they see your ID. You know, that is much more akin to what an online age verification system does. These bytes are transmitted over the Internet. That's inherently unsafe.

And despite what my colleagues may have or my colleague at AVPA may have said, the French government has not approved any age verification technology because none of them meet their privacy standards.

When you. I mean, I've been known to go to a place that sells alcohol in my life, and they. Some places, you know, especially late at night, they scan your ID and device that's connected to something on the world Wide Web, I would imagine. And I assume you're opposed to those devices as well, or there's a slight.

Difference in that alcohol consumption isn't protected First Amendment activity. And so someone who got access to somehow that you had scanned your ID to buy an alcoholic beverage.

Consuming a legal product is not a First Amendment activity.

It's not? No.

Or going to a.

Technically, legally no, it isn't. It's a little bit different.

We had an amendment to the constitution.

But also, depending on what you drink, I don't know. But it also may not reveal preferences you wouldn't want someone else knowing about. Maybe something that could be used against you in a divorce petition. Maybe something that could be used against you at your job, whereas your activity online might be a little more sensitive.

Yeah, I understand the, you know, the data concerns about IDs and things like that. I literally was just in another Committee testifying on a Bill about senior fraud, and I proposed this Bill, and unfortunately, and there are people out there. I don't know where, I don't go to these dark web websites. I don't even know what that is, but I'm told it's out there. And people have. They're able to accumulate all this information on people, Social Security numbers, things like that.

And so I don't necessarily want to make that problem worse, but presumably there is access to that information for a lot of the fraudsters already. And so I don't really like that, personally, the First Amendment argument, because it's not the. For my not being a lawyer, I'm a policymaker and definitely not a lawyer, but, you know, my children, and not saying that you are, they don't have a First Amendment right, you know, to do anything in my house, much less look at pornography.

But so, so there's protection there. I want to protect the data, and I think I would imagine the author, and correct me if I'm wrong, you would be open to taking some kind or continuing the conversation, at least that if there's methods to protect the data, you would be open to that.

As stated earlier, yes, I will.

Okay, great. Well, I wasn't here.

Oh, that's right. You weren't here.

So. All right. Because I have. I've. I've witnessed these new technologies. They're coming up all over the place. I just happened to travel to Europe, where they. Various things, they wanted you to scan your ID, and not necessarily just for travel, but they're processing the data locally, and I don't know how that works. That's way above my pay grade. And so I think there could be some level of trust, some kind of certification of the technologists that your information isn't going to end up out there.

And I don't know exactly if that technology exists but I think the premise of the bill to me is a supportable one, which is you should have to verify that you're over 18 to look at pornography and you can look as much when I say you. Sorry, I don't, but anybody can look as much pornography as they choose to.

And I'm very interested in the data protection and I would be willing to join in on conversations with the author on, on how we maybe firm that up a little bit, but because I'm concerned about that, but the policy itself I support. And so how do we make a better product? Let's do it, you know.

Thank you Mr. Vice Chair. Miss Ortega.

I just want to, I mean I want to thank the author for bringing this forward, but also being open to the conversation. I too had a lot of concerns and appreciate the fact that you, we're willing to take the amendments today. You know, as was mentioned, you know, throughout today, this is a very technical, difficult conversation that we're having in all of these bills. The technology is moving a lot faster than we can all catch up with.

But as a mother of a daughter, you know, I wish we had done a lot more when it comes to cell phone use and what we're seeing today and the harm that it's causing our children. Had we thought about some of these things ahead of time, we wouldn't be seeing that today.

And so while I respect and understand the concerns the opposition has, I also appreciate the fact that you're willing to continue working with us as I will be supporting today, but look forward to continuing this dialogue. It is a difficult one to have, but we need to protect our children. And I think ultimately that's what you're trying to do here today.

I agree. Thank you.

Thank you Miss Ortega. Mister Hoover.

Thank you, You know, the opposition brought up something about an alternative policy. I just wanted to ask like what is, what would be an alternative to this policy?

So if we're going just by what the Supreme Court said, it should be focused on parental filters, which have actually become incredibly effective. There are some issues with being over inclusive and so our organization has worked on model legislation that would help sites that essentially understand that they don't want children on them to get rid of them by using the filters. It's called a mature content tag.

It is very easy to add to any website and it would say if this user is under 18, they're blocked. I believe META actually came out with a framework that could work for apps and websites wherein your cell phone knows how old you are and it knows how old a child is if it's owned by a child and being able to not necessarily share a date of birth because that's sensitive, but at least have a flag that says this is a minor. Our websites would have no problem rejecting that traffic.

So you're saying essentially using the hardware.

Correct.

As the method of.

For software on the hardware, but essentially.

Kind of, you know, essentially like your device knows your private information versus.

I would imagine it's more like turning away a person at a bar at the door.

Yeah, I appreciate that. Look, I really appreciate the conversation. I think I actually had a similar bill last year. I do think that to my colleagues comments, I think that as a parent, this is just too important of an issue. But I do look forward to continuing conversations on privacy, specifically because I do think protecting people's privacy is important. But we'll be supporting the bill today. Thank you.

Oh, go ahead.

Thank you, Miss Dixon and I have one more question over to Miss Wilson.

Thank you to the author. I definitely understand the intent in terms of protecting our children underage against things they shouldn't be allowed to view, things that would harm them. I mean, think the actual language in the bill, things that are harmful to minors. And I think we, not only as parents shoulder that responsibility, but we as a community. I think this is one area where the government has interest in protecting children because, you know, not to be cliche, they are our future.

And so if we don't protect them when they're young, we can't be surprised at the kind of future we have in the future. Some of my concerns do relate to the First Amendment, like balancing that part, because the government interest also has interest in protecting our rights and very much strongly. There's a First Amendment right here. And so some of my questions are to your witnesses in following up.

There was an example given in terms of it's similar to going to a store and prior to purchasing goods that are above the age of 18 or 21 or whatever, showing an ID, which is definitely not the same, as was noted from the opposition, as retaining an ID. And there was conversation about tokens that you can have that I guess are on the computer or device that would be allowable for other sites, but that feels like a retention of sorts. So can you explain. Explain that to me.

Yes. So the tokenized solution would, which is something that the Spanish Data Protection Authority is very keen on, is having an application on your phone and you prove to the application, perhaps using a digital identity or using a piece of ID, that you are over 18. And then all that application does is send, effectively, a signed certificate that you are over 18 to any website that.

Is age restricted, so your data isn't sent to that person, just a green light, so to speak.

It's a green light which has been signed by a trustworthy authority so that the adult website can trust that when it sees that green light, it knows the only people who could have done that were people who have properly proven their age and are actually the person using that device. So it's not a month later where it may have been handed down to somebody's kid when they upgraded their phone. And the issues we have with that sort of secondhand phone problem, and one.

Of the things you noted, and I believe it was brought up in part of the other conversation I had it for here, was talking about France. Right. And they. It says in the analysis that it found that no method has the following three important elephant elements, sufficiently reliable verification, complete coverage of the population, and respect for the protection of individuals, data privacy and their security.

Yes. So what the Kinneil did, which is the data protection authority, and the l in Kinneil is for liberty. So they are very keen on freedoms. And the protection of privacy is. While they criticized existing methods a couple of years ago, they didn't just criticise, they came up with a better way of doing it. And what they come up with is something called a double blind solution. And that allows, again, through this concept of signed certificates, the Kinneil themselves have invented a solution.

They are satisfied, meets all of those requirements. So what I would say to this Committee, and as we go through the process of discussing the detail of this Bill, you can set the bar as high as you like for privacy protection as the CNIL have done, and the industry will step up. So there are now at least five providers in France who have developed solutions that meet these very new, very high requirements from the CNIL.

And so are you saying that the analysis noted that comment from 2022, but it's 2024, so since 2022, and that comment was made that there are. There's a gold standard now that satisfy those three and that that is an outdated comment.

Well, the CNIL themselves are the ones who've invented the solution to their problem. So they have. They didn't just say, we don't like the way you're doing it. They said, we don't like the way you're doing it. We've got a better way. We would like your industry to try and do it this way.

And we announced last week at a big conference in Manchester and England that we were, as an industry, moving to this double blind approach where it's impossible for the website to know the identity of the user, and it's impossible for the age verification provider to know which websites the user is going to. You know, I have a lot of sympathy.

I half understand what you said. I just want to confirm in the Committee analysis that we're reading, there's a statement from 2022 that had three components, that there was nothing existing at that time, that had those three components sufficiently reliable verification, complete coverage of the population and respect for the protection of individuals data and privacy and their security. And although that came out of France, I would say that that is a standard that we have in the United States.

Okay, so we've established that they've made that statement. It's put in our Committee analysis. And so what I'm asking is, and you were talking a lot about CNIL. That's the CNIL, the France for people in the room, National Commission on Information Liberty. You were saying. You were saying that they made that statement in 2022. However, today on what is it, the year of our lord? April 16, 2024 they have now met a standard of those three.

Indeed. They went and worked with a Professor, Professor Blasey, a French Professor in cryptography, and he worked with them to come up with this new solution. What we've been subsequently trying to do is figure out how we commercialize that because.

Okay, so a new solution exists. It is yet not deployed.

Well, I mentioned that five French providers have stepped up to provide that. And the official regulator, Arcom, which is in charge of Internet safety, works with CNIL. And they've just launched their consultation on the detailed rules that require at least one method for any adult site to meet these requirements of double blind that the CNIL have said. So what we're seeing around the world is regulators raising the bar to, you know.

To meet the three. We're going to meet the three standards. But it is not a. It's not something that's active. And I do get in technology, things are exponential. Something cannot be active today. And literally a month later, it's all over the globe and it's completely active.

Yeah. So one of those five suppliers is a member of my trade association, and they currently offer that as a product in the market today. People are actually using that double blind approach.

In the market today.

in Europe, but not, but not in the US yet.

In the market today.

No. I mean, obviously there's no law here requiring it, but.

Right, right. But it is being actively. It's deployed. Someone is using it.

Yes.

Right. And part of the association. And I do recognize that you are testifying with the hat of the association. And so, of course, there's been a, not to say there's no, the author has this motive. In some ways, it benefits because you are part of the association that provides this basic service that we would then be putting into law. And so recognizing that, at the same time, recognizing that.

And so I don't mean that in a disrespect to the author in the same level, we need to protect our kids. So how do we do that? I'm very much concerned about the privacy aspect of individuals, especially as it relates to this particular topic, because it can be weaponized against the person.

If there is some type of, and I'm not talking about minors, I don't want any child looking at these sites in particular, as it's laid out in the bill and the definitions that Madam Chair has given. But I also recognize that we live in a society where people weaponize us all the time for every little thing we sit in privacy.

You know, I'm sure after this committee today, there's going to be tweets about various decisions, and you must like this, or, you know, whatever we are, and then just imagine if that's, someone was able to get that data that, because of that type of verification system, because our, this, the language in this Bill doesn't require the type that's used.

It just requires an age verification system or some type of system, system that, that could be used against someone in a nefarious way if it's hacked or whatever. And so we live in that world.

So I think, as you know, if there's support for this to move forward, I think you have to take, because we're in privacy, right, and consumer protections, you have to take that into consideration because there is a role, I think, for the parent, there's a role for devices, as my colleague has noted, you know, putting things on, or I think opposition noted in that conversation, that you can do things locally to be able to prevent that, not necessarily on the burden of the site where you're giving third party information.

I think that's, you know, like I said, just balancing the needs of everybody else who has an absolute right to be able to exercise their free sheets, to produce content, and exercise their right to be able to take part in it by viewing it. And so we want to make sure we're balancing those needs. Thank you.

Thank you, Miss Wilson. Miss Dixon, did you have another question?

Thank you. Madam Chair, just a few follow up questions. So I just kind of wanted to distill everything down. I think at least I, this is where I'm coming from. It's not a good thing to have children, youth under 18, accessing pornography. We're trying to find a way in this hyper technology world to find a way to make sure we protect everybody's rights, or at least, I guess, the children's rights. I don't know.

I don't like my colleague here saying I didn't know that children under 18 had rights to access pornography in a free speech context. Anyway, this seems to be an obstacle to me. I think the larger issue is how we, the technology changes so quickly, and just these protections that we've just heard are satisfactory to me.

And technology is moving so fast that if we don't try to put a stop to this kind of material being accessed by children on the Internet, and we've heard earlier in the testimony of, of the 25% of the world is accessing this information, we can't do anything about adults, and the US Supreme Court has dealt with that decades ago.

But as far as children, I think we err on the side of doing whatever it is to protect our children and move forward with that concept that these technologies can protect the identities of people, people, children, whoever submits there through that process. I mean, we've all used those kinds of things. I mean, even with your health plan and medical plans, you submit copies of your driver's license. That happens, and then it disappears anyway. I'm in strong support because it's time we do something.

We could debate this to infinity and beyond, but we have to protect our children, and that's very simple to me. In my opinion, I will be enthusiastically supporting this. Thank you.

Thank you, Miss Dixon. Anyone else seeing? No other Members with questions? I want to thank you, Assembly Member, for your partnership on this. I know that the original bill is introduced, probably, as evidenced from the Committee amendments we had concerns about, but I shared your goal here.

There was an article in the New York Times just this weekend talking about the alarming number of college age girls that are reporting being choked while engaging in sexual activities without their consent and the dire physical ramifications of that on their brain and the oxygen that is deprived to their brain. If I recall correctly, it was 40% of college aged girls.

And the studies cited discussed exactly what your witness talked about, which was that this was most likely a result of very young kids being exposed to very violent pornography, which I know is not something that the opposition wants either. But it is something that is happening, and it's happening in a lot of cases, as was noted by the proponents of this Bill, accidentally at first. And I know that as a mom, you know, it's then sought out.

I'm not saying it's not also sought out by youth, but I know as a mom whose kid has written reports and while researching for those reports, has definitely accidentally encountered websites I don't want my child on. It happens to all of us parents. There's no question about that. And we use all the parental protections that are available to us.

And so, you know, I do think that there is no question in my mind that making it more difficult for our youth to access these sites is in the best interest of everybody, whether it be the person viewing it or the person who they will engage with in sexual activity later in life. And so I think this is a really important policy.

And I think that we have done a lot in this Committee, and huge kudos to the staff for their work on this, to really narrow it to what is constitutionally reasonable, to make sure that it passes constitutional muster, which is always in the best interest of the State of California so we don't have to spend a lot of money on litigation. And also to protect privacy, as was noted by the proponents. We did say, you can't keep this private information. That was something that we added.

I also think, though, there is, if we can match the Europeans in any way as it relates to privacy, we are winning. The Europeans privacy laws will eclipse anything here in the United States any day of the week. And so to the extent that the Europeans have created better standards for age verification that are more privacy protecting, I think that is something we could look at.

We do a lot of legislation in this body that mimics what has happened in Europe, and it's something that we should consider if they do have a way of doing it that is more privacy protecting than is what is currently in the language. But I just want to thank you, because I do think we have a serious problem here, and I think you have been open to really addressing it in a surgical way, but also doing what is in the best interest of our kids.

And I said this in a Committee hearing earlier today, but I'll say it again, I truly believe that as the elected Members of the State of California, our number one obligation sitting in these seats is to protect our kids. And so, you know, I think that this is a step in the right direction. I thank you again for taking the amendments and working with us. And I know you will continue to do that, because you've already proven that you will. So with that, would you like to close?

Well, thank you, Madam Chair, for those words. And yes, rest assure, obviously the sponsors and myself, we're here to bring policies that will protect our children. And I think with the Committee's help in narrowing and making sure that this is within the laws that have been set forth, we'll make sure that this happens. As far as opposition, please come to my office. I'm sure we'll get some cards for you guys as well. I'd like to hear from you as well as we've already heard from others.

With that, I would respectfully ask for your aye vote. Thank you.

Item number five AB 3080 by assemblymember Alanis. The motion is do pass as amended, to the Judiciary Committee. [Roll Call]

Bill has nine. It is out, but we will leave it open for absent Members. Thank you.

Great. Do we have a motion? I don't think we're gonna moved. Move. 2nd 2nd Madam Secretary, please call the roll.

Thank you.

I believe Ms. Irwin is now presenting for Dr. Weber. Is that right? Are you? So then, Ms. Wilson, if you want to go next, you're up next. Okay. Is that okay? This is AB 3138. Ms. Wilson, when you're ready.

Thank you. Madam Chair, Members, good afternoon. I am thankful for the opportunity to present AB 3138 in front of the Committee here today. I would like to begin by accepting all of the Committee amendments listed on page 8 and 9 of the analysis and thanking Committee staff for their work on this measure. Members, building upon a successful pilot program launched all the way back in 2013, AB 3138 will provide consumers with the option of choosing to purchase a GPS enabled digital license plate.

Under current law, when purchasing a vehicle, consumers are typically given a metal plate. However, if a consumer wants to choose to purchase a digital license plate that is approved by the DMV, they currently have that option. AB 3138 would provide consumers with the option of choosing to purchase a digital license plate that is equipped with GPS. These digital license plates typically cost around $500 and require a subscription.

This bill also puts in place privacy standards for the GPS plate that far exceed protections in current law for other comparable GPS equipped devices. AB 3138 requires the following protections for all GPS enabled plates sold in California. First, the GPS shall be capable of being permanently disabled by means of a non-reversible method that ceases all functionality and traffic, sorry, tracking information capabilities. The GPS shall be capable of being manually disabled and enabled by a driver of the vehicle while that driver is inside the vehicle.

The method of manually disabling and enabling the GPS shall be prominently located and easy to disable and enable without requiring access to a remote online application. The method of manually disabling and enabling the GPS shall not require a password or any login information. And finally, once the vehicle location technology is manually disabled from inside the car, the only method of re-enabling the technology shall be manual from inside the car.

The registered owner of the license plate, the manufacturer, the DMV, or any other entity shall not have the capability to re-engage the GPS through any remote means. These protections far exceed protections in statute for other GPS enabled devices such as the cars themselves, apps used inside the cars, our cell phones, and countless other legal products available to consumers at much cheaper prices.

While GPS, like other technologies, has the ability to be abused by bad actors, the appropriate policy response is to put guard rails and protections in place like we've stated earlier, not to simply ban the technology in one small product while allowing it to exist without recourse in virtually every other comparable product. With me to... Excuse me. With me today to present in support, I have Neville Boston, the co-founder of Reviver.

Good afternoon, Chair and Members. My name is Neville Boston. I'm the founder and Chief Strategy Officer for Reviver. AB 3138 builds upon legislation that has been voted on by the Legislature previously, AB 984 and SB 806, which was the original pilot that was passed back in 2013. We've been working with the DMV and CHP over the last eight years, and the company's been around since 2009. So we've been testing this technology for the last 15 years.

This bill clarifies that digital plate must comply with all provisions that to ensure privacy of the consumer is not compromised, while continuing to strengthen the benefits provided to businesses. To be clear, we offer two plates. The plate that I have here is a battery operated plate, which is about 95% of all the plates that are on the road. It has no GPS at all within this plate. Again, this is the majority of the plates, and they're opt in.

We have a wired version that has GPS that is normally used for fleets, but we believe it should be available as an opt in. You know, not an opt, you have to opt in to it. It's not provided to you for consumers as well. Reviver looks at the services that we've built into the plate to help protect and to make it easier for them to do different DMV services.

The language of the bill clearly states if an alternative device fails to meet the requirements, vehicle location technology shall not be included in that device. As AB 984 moved through the legislative process in 2022, the amendments that were taken were a result of conversations in regards to having a kill switch that was built into the plate. We have done that. Digital plates and geolocation technology is beneficial for fleets.

We have several fleet customers both local and in and around the state because we're legal in not only California but Arizona, Michigan, Illinois, Georgia, Colorado, and Texas for commercial fleets. So we've been working hand in glove with them kind of across the board. In addition, we have a law enforcement liaison that's been able to train about 35,000 law enforcement professionals across the country.

Thank you. Anyone else here in support of this bill?

Margie Lie, Samson Advisors, on behalf of the California New Car Dealers Association in strong support.

Thank you.

Dominic Di Mare here on behalf of the California Police Chiefs Association, also in support. And find the kill switch addition a good feature for personal safety.

Thank you. Anyone here in opposition to this bill? Come on up, please.

We didn't coordinate beforehand.

You're too nice to each other.

We're working it out.

No conspiracy. You can see.

I know we should have compared talking points. All right. Hi, Madam Chair and Members. Samantha Corbin with the Electronic Frontier Foundation, respectfully in opposition to AB 3138, which would turn a vehicle registration product into a surveillance tracker that records everywhere the vehicle goes. This is bad news for Californians everywhere, but it's especially concerning for domestic violence survivors, the LGBTQ community, immigrants, people seeking reproductive health care, rural communities, and communities of color more broadly. Numerous domestic violence, LGBTQ, reproductive justice, youth, and privacy organizations raised serious concerns with AB 984 in 2022.

That bill also proposed allowing GPS tracking devices in passenger vehicle digital license plates. AB 3138 would remove the language in that measure that was negotiated to address these concerns, though all of the concerns raised by those groups about the harmful impacts of GPS trackers and passenger vehicle digital license plates remains true today. Creating these tracking devices would jeopardize confidentiality and increase the risk of stalking. That's why numerous domestic violence organization joined privacy advocates to oppose AB 984 until it was amended to prohibit their inclusion.

People who abuse have no incentive to let their partner know that their car has become a tracking device. And even if an individual is suffering abuse and is aware of a GPS tracking device in their vehicle, they may have no choice but to allow for it, or they may not know how to disable the tracker. Regardless, it can put them at risk of further violence. For example, a woman was recently killed in Antioch after being located through a GPS tracker installed on her vehicle.

Domestic violence survivors are not the only people put at risk by this measure. It could jeopardize the safety of someone traveling through the state from another state that criminalizes abortions who is not aware that their vehicle or ride share is recording their GPS. This data could be used against them. ICE could also use GPS surveillance technology to track and locate immigrants, as it has done with other location tracking devices.

Unsupportive parents of queer youth could use GPS loaded plates to monitor whether teens are going to local LGBTQ centers. Finally, there are security concerns. Shortly after AB 984 was signed into law, researchers uncovered that Reviver, the proponents of the measure, had an alarming security vulnerability in their systems that made it possible for infiltrators to track GPS in real time and even change what was displayed on the plates.

If a company producing these digital plates for passenger vehicles cannot secure the location or the data, promises that it can protect sensitive location information for vulnerable people ring hollow. While acknowledging that continuing to prohibit trackers and license plates does not fix every problem with vehicle tracking, we do not believe that their inclusion helps the situation.

Rather than remove the language that allowed all groups in opposition to AB 984 to go neutral, AB 3138 could instead address additional problems with the code sections that our groups have outlined in position letters previously. Thank you for the opportunity to speak to you today.

Thank you.

Yes. Oakland Privacy in respectful opposition. As my colleague mentioned, mere months after digital drivers licenses were made available for sale in California, hackers did penetrate Reviver's server. We want to be clear, they were able to label themselves as super administrators within the server, and they could remotely update, track, and delete any digital license plate. Let's be clear, if DMV had been hacked in this way, all heck would have broken loose in the state. Heads would have rolled. Sending people to unsafe products is bad.

But adding on sensitive location data to a product that has already demonstrated some level of lack of safety is, in our opinion, kind of crazy. The bill's premise is that this is a question of consumer choice. We just want to be clear. There is no other GPS enabled license plate available for sale in California. The company that is here is the sole provider. It's a monopoly.

And it's a question here of sort of, you know, to give some people a choice, are we taking choices away from other people? Gay teenagers told not to go to an LGBT center, 16 year old girls trying to get contraception, a battered wife trying to visit a supportive sibling. If you are in a situation where there's only one car in the household, where there's not so much public transit, and when there is a level of coercive control in the household, it's not necessarily a choice.

So I just want us to think about whose choices we're sort of prioritizing here. And in those kinds of situations, you turn off the GPS, you're going to get a beating when you get home. I mean, that's how it works. So it's not just a question of choice. It's not as simple as that. So we don't think there's any need for California to sponsor this dangerous product. There's no market competition to keep the prices affordable, which, by the way, they aren't. It's, I think, $699 plus 250 a year indefinitely. There's only one supplier, and there's no clear demand for the project, and it just, it does not make sense. So we ask you not to support AB 3138.

Thank you. Anyone else here in opposition to this bill?

Good afternoon, Madam Chair and Members. Robert Herrell with the Consumer Federation of California. We don't have an official position on this bill at this time. I want to make that clear. But we were part of the coalition that worked with the Assembly Member in 2022 in what we thought was a reasonable negotiation. So we're concerned at part of that direction that this bill is going. It's still early in the process, and it seems to me there would be some time to figure this out over the next few months. But we do have some concerns here. Thank you.

Thank you. Seeing no additional comments in the room, we'll bring it back to the Committee. Mr. Vice Chair.

Thank you. Some questions for the opposition, Ms. Corbin. Actually, first, you know, I was recently, I keep talking about podcasts, but I was listening to a podcast about domestic violence victims being tracked by some cars, frankly, and that being put in. Yes. It was also the New York Times, sorry, everybody knows my running playlist now, but that was shocking to me. Now, we've seen some legislation, I think, in this building on that particular issue. Oh, the next bill we're going to hear. Good.

But I have obviously a ton of concern about, in that particular situation, the spouse couldn't turn off, with the car manufacturer, couldn't turn off the reporting of the data to data brokers, essentially. But in some of the instances that you were discussing, like the tracker on the car and things like that, those are all not done in those cases without, with the consent, somebody put a GPS device on a car, for example. In this case with the license plate, an individual is going to know they have a license plate on their car that's a digital license plate. I concede that. There's a chance they might not know that there's GPS in there. I concede that.

But there are slightly different situations there where, where one, you knew about it, one, you didn't know about it. You knew you have a product, you didn't know you had a product. So is there any circumstance, because I have sympathy towards those points in particular. I definitely understand them. I don't want people being tracked. But is there some way we can get to a point where we can have this product out there and your clients would feel comfortable?

I appreciate that question, Assembly Member. I think that's a pretty significant challenge for a variety of reasons. One, we find with tech products, as they sort of proliferate the market, oftentimes they become not opt in because they're everywhere. Right.

The traditional alternative evaporates, and you no longer can get something that doesn't track you. So those type of things are very concerning if this becomes the status quo on every car, for example. But also, I think Ms. Rosenberg brought up a great point, sort of building off on that domestic argument. If you are in a single car household, perhaps you are not the person who made the car purchase or consented. You might be a spouse that's disempowered. You might be a child in an abusive relationship.

And so your option is to ride in the vehicle or to drive the vehicle and have someone have access to where you are. And so there are a variety of situations where there may be a choice, but it, in fact, is really not a choice. It's not as clear cut as being able to toggle off your location from an abusive partner. And those type of scenarios, we believe, at least at this time, so significantly outweigh the benefits of this product that I don't see a series of amendments that allow for this use in personal residential cars that would remove our opposition.

Yeah. And I want to be very clear, not dismissing these concerns at all. I just think of the number of products out there that track individuals, whether it's your phone or your car, already, that are tracking you or devices, you know, the, you know, the airtags or whatever you can put anywhere.

So it's just kind of interesting to me because I think what I'm hearing a little bit is, hey, we can't have this product with GPS in license plates when there are all these other products out there that are doing that, that are actually more difficult, I think, to have the kill switch.

And so if you don't mind, I'd like to transition to the sponsor, which I will say is, in my district. I never met or we didn't ever discuss this issue until right before this Committee hearing. And so I appreciate the opportunity to talk to you, but I had expressed to you my concerns with that and what would happen. And I think there's language in the bill for this and maybe even your terms of service to prevent that. But really, when we think about policy, we have to think about there will someday, despite you being maybe the only entrant into this field, this law potentially will open up other competitors of yours.

And just to note that it already has. So he's not a single. There already is.

They're starting.

Yes. There's a process. You have to go through. He is the only person that has completed the process. Let me say it that way.

Okay, great. So in law, we want to make sure that this kill switch applies to everybody, not just maybe the good work that you're doing, but can you explain the process if maybe an abuse, abused wife or spouse calls you and says, hey, look, I'm really concerned that I'm being tracked with this license plate, and maybe the car is in the husband's name.

Okay. In the event that that plate that's on the vehicle has GPS in it, once we have the conversation, we would send out kind of a kill message that would basically sever the GPS within the plate, make it non-operable. So they wouldn't be able to track. There's no more information that's given, it's killed. We actually literally worked with the chip manufacturer and, you know, our team to verify that that was done. And that was part of the conversations that we were having with folks from the DV community around ways in which we could help protect them.

And I would just add on is technically they don't have to call you. You noted in your just now a testimony call you, but per the bill language, what would they do?

They would be able to switch it off within the vehicle and basically kill the ability for the plate to connect to GPS.

Is there a... When you get one of these license plates, do you let the DMV know I have this digital license plate?

Yes. So we're connected to the DMV. Anybody that gets a plate has to have the plate provisioned. So they go through the same process, registration information, validating that, and then coming up with the number letter configuration, and then downloading that information to the plate.

Okay. The only reason why I ask is because in domestic abuse situations, they're often desperate circumstances for help. And so being able to even rip off the license plate, and if you get pulled over or something, say, hey, look, this is why I did this. Right? And it sounds like that could be verified that it was one of those digital license plates. But... Well, I really appreciate your approach to this issue because it's really important.

And I know the author, obviously, with some of the concerns that were raised, those are all very important to you, and I know you'll continue to work on this particular issue. And I don't know if we'll ever get to the point where we're making the opposition feel comfortable, but I know that that's important to you, and I appreciate that.

And again, the sensitivity of the issue the entire time from the manufacturer I think is a really important testament statement to you. Because it is a really important issue to me, and I think everybody on this Committee. So thank you.

Yeah. And we'll continue to work with everybody to make sure that we have the best and safest product on the road. We've never been or shied away from having tough conversations. I would say, and I'm not sure a lot of people here may remember, but Joe Simitian, who was a huge privacy hawk, was actually the first person we talked to when we first were thinking about running legislation back in 2009.

So, from the beginning, we've been having these conversations, and my former business partner was good friends with the folks that started the EFF, and we actually had conversations with them back in 2009. So it's been 15 years, so it's not a situation where we've ever shirked away from having tough conversations. We want to have the best information possible at the table, and we've always been open to, you know, making the product better across the board.

And I would just note that even in 2022, people talk about the negotiated deal, that we negotiated this out. And what it was is that we offered up what if we did XYZ, but we couldn't do... The technology, so to speak, wasn't there yet.

So we took that language out to be able to meet those objectives, knowing that we would come back separate so we can complete the pilot and then talk through what can we put into law to allow this license plates to be as safe as possible and recognizing there are individuals who may opt in to having GPS but at some point may no longer want to have GPS, whether that's temporarily or permanent, and giving them the full capability of being able to make that decision, whoever they are, as long as they're in the vehicle, so to speak.

And I do wanna commend you, as the author, for being... I mean, you put survivors first in this legislation. And the kill switch, which now you will see has been added to the other bill that will follow, really was... Came out of the discussions we had because you wanted to make sure that they were wholly protected. And so I think that I want to commend you for that.

I also want to add one thing because I know you are a leader in protecting LGBTQ individuals in accessing care. You are a leader in protecting women accessing abortion care in our state. So you would never in a million years want to do anything to risk those folks coming into California. So I guess I will also just put on the table that we do, that I have passed legislation that ensures companies don't turn over that data.

And there's no reason we couldn't add that to this piece of legislation and require the folks that sell the license plates here in California never turn over data related to healthcare access, which I...

Absolutely.

I'm sure you'd be amenable to.

I thought we dealt with that one, but if it is not in there very clearly, happy to add it before we get to the floor. Because there's no way, shape, or form do I want anyone to be able to track this data, whether... except the person themselves, who's opting into, the single person. Whether they are another company or even the government. I think that it's private individuals' information. They are the ownership of that data and should not be shared with anyone, especially if it threatens their safety or their ability to live a free life.

Yes, absolutely. Thank you. Mr. Bryan.

Yeah, thank you. One thing, the author. I know that this is a pilot that's been extended multiple times, like three times. Obviously, just naturally have great privacy concerns, and only more greatly affirmed by law enforcement coming up in strong support. Correct me if I'm wrong, in the previous pilots, there was a requirement that the device display a visual indication in the location technology, if it was an active use, if somebody was actively tracking you, there had to be some sort of visual representation of that. And that's been removed in this bill. Is there any particular reason why that's the case?

I believe in the original, it was on the app itself, and so the only thing we had, and it didn't exist in the car before, it's the new ad. So when during the pilot period, the everything was available, GPS or not, and it really was... The battery doesn't include GPS at all. And so majority people chose the non-GPS option, so there's nothing to indicate.

And then in the fleet version, that the wired in it would be in your app would tell you whether it was on or not. And so there wasn't ever like a visual, there wasn't anything in the car. Now we've created something visual in the car through having it wired in all the way to the console where you can kill or turn off.

And I think that, I'm sure if... I don't want to answer for the technological people, but if it means that that needed to be lit or something to indicate that, so you can see it. I know dashboards, a lot of very busy right now.

But I know that there's an ability to be able to see, and I think we have an identifier on the vehicle that you're entering a car that has GPS, but that's just a broad, that this is GPS enabled, but that's a broad disclaimer. Not that it's active or not on, but just knowing that, hey, you're entering a car that has this capability. But I'm happy to make that more safe or make that more apparent to the driver themselves.

I think to note from my colleague's comments is that by virtue of having the plate gives you some indication that you may be entering, that this might be a tracked vehicle because it's not, you know, it looks very, this plate looks very different than a metal plate. So you recognize that you're in something that's different.

Gotcha.

Oh, yes. Mr. Dixon.

A couple simple questions. I noticed that the language changed with the CHP, the California Highway Patrol, are they part of the approval process, or they're not definitive, or what is their role?

They're a part of the approval process through DMV. So the ultimate sign off is DMV. One sign off, and then CHP is a part of the decision making process, but it requires only one sign off. Because the, originally, when they were doing the pilot program, then you had to get a sign off from CHP and DMV.

But now that we have all the rules in place, all the laws in place, they've worked out, I don't know what the right word, technology, what the right word is, but I guess the acceptable specs they were tried out with, at the time, the only company who had filed an application. And so they no longer need their own separate sign off, but they are a part of the sign off for DMV.

So there's only one agency that checks the box to say... And not them, because they've already been through that process, or is for any other company that's through the process, or them if they want to change the plate. And that was CHP was fine with that. They would...

The Association supporting this?

Yeah. Well, I wouldn't say supporting the bill. They've given all their technical... We work with them in 22, the previous author who did the pilot work with them, they have signed off on every step of the way. So I don't want to say that they're, they haven't put a letter of support. They're neutral on the bill, and they've given... Every amend that they have asked for over the process that I've been with it for two years, and I know with the previous author, previous to me, they have, all their amends have always been taken at face value.

All right, another question on the making of the digital license plate. Is it reflectorized?

Is it what?

Reflectorized material currently on all of our license plates. It's reflective material at night with the light shining on it.

Okay, so the current license plate is retroreflective. Ours is reflective. So light shines and it shines back. It's not retroreflective.

Okay.

You can see it at night.

You can see it at night.

And just as a reminder, this came up in our other Committee. If, for whatever reason, the power, because it is battery operated or wired in, if there is a loss of power, the default is visual, where you can see the plate clearly, so it doesn't turn off at night.

Yeah, night, every day.

Although, I'll remind the Committee that our jurisdiction is privacy. That question actually is in the jurisdiction of the Transportation Committee.

Thank you. All right. It's one other thing I was going to ask. Oh, about... Well, it's just in terms of the procurement process, will there be other competitors or participants in this marketplace?

Absolutely.

Yeah. There are other... There are other applicants who are in the process. Whether they complete the process or not, we're aware of others. So it's not just... So although it is one person in the market, I wouldn't call it a monopoly in that sense. They're just the person who... They have only one person who's completed the entire process and has an approved plate with the state. There's nothing in this... This is a digital license plate bill. It is not Reviver's bill. So any...

So this is not exclusive to Reviver, gives them no exclusive rights. All of the language in the original bill, including this bill, does not give them any favor in comparison to their competitors. It's just they are the only person who has completed the entire process and has an approved plate.

Under their specifications?

Under their specifications are consistent with the law. And anybody else who completes that process would have to be consistent with the law.

Are they patented?

They're patented. I'm sure their technology is patented.

Participate with that, they'd have to come up with a different type of product.

I guess technically. They couldn't use any of their patents, but they could get their own electronic version of the license plate. They could get their own digital license plate using their own technology and meet the requirements of the original law plus this new law. That would be the requirement. They would have to go through those standards.

Let me, just really quick, AAMVA, which is the automobile administrators for the US and Canada, all North America, they set best standards. We've been working with them for the last 15 years. We have a standing meeting with them every month to talk about best practices and technology. So the entire time we've been kind of working hand in glove with them, with, also with law enforcement to make sure that whatever product we put on the road meets their standards. Their standards are 75ft during the day, dusk, and at night. We've met all of their standards for visibility across the board. And no, this is not Transportation.

Yeah, that's a whole other...

But that's another Committee.

All right. Thank you very much.

Yeah. Just want to make sure.

Yes, and whatever and whoever applies will, because of the laws we put in place and potentially more, will make sure that it is safe.

We trust the Transportation Committee, Madam Chair, is taking care of that. Mr. Hoover.

Very briefly, I just abstained on this bill in Transportation due to some of the privacy concerns, and I just want to say thank you to the author for continuing to work on it. Thank you for the amendments. I do think those have gotten the bill much closer to a place where... Just to a much better place. And so I will be supporting the bill today. Thank you.

Thank you.

Look at that. Ms. Ortega.

Thank you. I had a quick question. Just, I think I already know the answer, but I want confirmation that nothing that we are removing in terms of, you know, there's been a lot of privacy concerns that have been raised. And we're removing a big piece of that, and so there's a lot of hesitation. So I want to make sure that there's a provision on employee use, you know, their employers do not use this information to be able to follow or use that against an employee. So I want to make sure that that has stayed in place. Is that correct? No?

That's in a, not in this bill. So they, in a separate bill, as it relates to, as I understand it, law is that employer cannot use your data. Because this is taking what is allowed for GPS to go to a personal consumer. But we already today allow GPS for a fleet.

Please don't interrupt her. Please let the author finish, and then we'll come to you.

Sorry, go ahead.

I was just noting that this bill expands it to personal. But as I understand it, there are separate laws around the use of GPS data, whether it's in a license plate or their phone or putting Tile or Apple in there about what an employer can use against an employee related to GPS data while that employee is working. And so it didn't need to be additional language in this bill.

But it remains in place. So it's already...

Yeah, it doesn't exclude it. And as a reminder, this bill expands to personal, so it doesn't really mess with the existing allowable by law for fleet GPS.

Okay.

You get what I'm saying?

Yeah, I'm just trying to make sure there's not a loophole, which is often what happens.

And so that's an area that I'm very concerned in and want to make sure that, you know, we can...

No.

Absolutely. Before we go to opposition. That's my goal too to protect workers and understand. And I remember when I first got here, there was a whole thing about using GPS and what workers were doing, especially as it related to remote, and supported those bills. And so if there is any language that currently... I don't think there is, but my laptop's over there.