well thank you very much for all being here this morning especially uh members of the legislature but all of you were very excited uh to have this press conference to uh welcome you to a really impressive legislative package that uh our legislative leaders are going to spearhead at a very challenging time my name is Drew lbert I am the director of the California Initiative for technology and democracy uh we call it cited um and we're very excited to engage now on obviously what's one of the toug
hest issues facing America today um my background actually was working almost 20 years in this building as you remember um yes as the chief counsil of the assembly Judiciary Committee and um a great way to do uh work after that is to try to help protect our democracy this program of cited is actually sponsored by the um California common cause which you all know is one of the preeminent democracy protection groups uh in the state and they got to understand that you know it's not just getting peo
ple to the uh voting uh locations or making sure they get ballots that in the age of social media Amplified disinformation if we don't deal with those issues um getting people to the polls isn't going to do the trick to protect our democracy um so I want to thank especially the staff of the members here today who have done just an incredible job in making this happen today um and I'm going to just uh show show you one quick video that's just literally less than two minutes that I think will nice
ly set the stage um for the comments that the members are going to make today about their bills and why they're so important so if we could start that Ken uh this is an interesting video that I think you'll find compelling I want to show you two video clips one of them is real and one of them is fake can you tell which one's real okay this one was kind of easy obviously this is the fake one like you can see the color on the face isn't quite right and hair is obviously like a wig but this video w
as like from a few years ago let's look at a more recent example look at these two clicks one of them is real and one of them is fake can you tell which one it is okay what about these or these which one of these is fake and which one of these is real maybe you're some Wizard and you could tell but I couldn't and my guess is that most people can that's because things have [Music] changed [Music] in the last few years we've crossed a threshold into a world where moving images are manipulated in w
ays that make them indistinguishable from reality this is our new world it is the world of high quality deep fakes a catastrophic potential to public trust and to markets that could come from Deep fake attacks this is happening quicker than any of us could have imagined so I want to get you up to speed on what's happening with deep face how they're revolutionizing entertainment just ID and creating new challenges and fears for the people who make our laws and the people who enforce those laws ma
ke it look like anyone is saying anything at any point ultimately I want to show you how much of a threat deep fakes actually posed and answer the biggest question of all what is even real imagine if manufacturing on them wow um I saw that and was really struck by it because what a question we're now facing in our society and in our politics and our government right um the question what is even real anymore and obviously what is truth we're starting to obviously lose a sense of what that can be
because of these new technologies so today the great policy makers in front of me are are going to be announcing their legislative package and I just want to set a little stage for that every day we are seeing headlines of these enormous risks that they are trying to address now just today in the New York Times was the headline house passes Bill to force Tik Tock sale because of the threats that are posed by that enormous social media company and the Massachusetts in its Institute of Technology
tech report had this headline this morning which some of you may have seen let's not make the same mistakes with AI that we made with social media so we just concluded one California's most competitive primary elections in years and we're now looking ahead towards what we know will be a monumentally consequential general election in the thick of what is truly our first AI election we don't know where it's going to go we know that deep fakes are going to continue to overwhelm our political discou
rse leaving voters unsure of what images audio or video they can even begin to trust and alarmingly just as this is happening we've all read these articles about the social media companies many of them saying that they've dramatically cut their trust and safety teams that are tasked with trying to prevent this disinformation sadly the American public is understandably not prepared imagine false audio of an election official caught on tape allegedly saying that their voting machines can be hacked
that's now easy to make in fact the first known attempt to use artificial intelligence in this election was was actually put together by a magician in New Orleans you may have read about who actually is the one who created that deep fake audio of of um our president allegedly telling Democrats not to vote in the New Hampshire primary it was very persuasive it goes obviously it's a bipartisan challenge uh Ronda santis of course did a deep fake of former president Trump um and that was a very uh
deceptive as well so we must act now um that's why cited is partnering with these great leaders to see if there are ways that we could at least make some incremental progress in slowing down the social media Amplified disinformation that we're all facing my final comments before I get to introduce our our our our panelist today um a few things I wanted to make clear about all of this legislative package that you all have put together so well these members are not trying of course um uh uh uh to
address all of ai's threats that's impossible so we are all focusing um on the effort to protect voters from disinformation and to protect our democracy everyone's fully aware of these first amendment constraints supporters of the First Amendment we we love free speech um so we are trying for transparency but certainly not censorship um um we know that they are pro-innovation and these bills have been crafted to ensure that we protect Innovation while at the same time trying to surgically addres
s these problems and we're trying to be as nonpartisan in this approach as possible because these aren't partisan issues um whether you're a republican member of the legislature or a Democratic member of the legislature if someone deep fakes you in a very unfair way it's going to be harmful and you're not going to have too much recourse to try to solve that problem so with it I want to now introduce just a quick uh um technologist uh that is very aware of these issues um and who's a member of th
e cited team um Smita Rajan is a council at a a very impressive technology firm Autodesk and just has a few preliminary comments if you could come up uh Smita that would be great and then we will turn to the key part of our conference which is the members talking about these bills good morning everyone um and thank you drew and all the legislators here today for giving me this opportunity to speak about this important topic um I want to begin by stating I am speaking in my individual capacity an
d not on behalf of my employer the rise of AI has undoubtedly made it easier for disinformation to spread raising concerns not just among those in the public and the policy makers but also those that are in the tech industry like myself what makes this moment in our history of tech Innovation so unique is that through AI generated deep fakes with fabricated news articles targeted social media campaigns and falsehood thank you falsehoods can now be disseminated at an unprecedented scale in speed
but something that doesn't really get talked about nearly enough is that disinformation is also bad for business the proliferation of disinformation poses significant risks to businesses as well as the community inaccurate or misleading information circulating online can harm brand reputation erode customer trust and ultimately impact the bottom line businesses are increasingly recognizing the importance of combating this issue not just for ethical reasons but for their own sustainability the sp
read of misinformation can create a challenging environment for businesses to operate in it often leads to Consumer uh confusion hinders effective communication strategies and creates legal and Regulatory Compliance challenges addressing these issues requires a multifaceted approach that involves collaboration between public and private sectors as AI capabilities evolve so do the methods use to manipulate information astroturfing deep fakes data poisoning impersonation of public figures and mali
cious use of AI assistance are all real threats no longer theoretical when it comes to addressing these challenges wellth thought out regulation plays a crucial role it provides ethical guidelines for businesses enabling them to design AI products responsibly bills that focus on authenticating AI generated content on social media platforms and detecting deep fakes are crucial steps forward by implementing such measures we can bolster the Integrity of our Democratic processes and protect the publ
ic from the harmful effects of misinformation contrary to the belief that Innovations uh regulations stifles Innovation well-designed regulation can actually Foster Innovation clear guidelines provide businesses with a road map for compliance freeing up resources that would otherwise be spent navigating uncertain legal Landscapes finally it is imperative that we align the objectives of AI with the best interests of humanity AI should serve to enhance Our Lives not undermine them this means empha
sizing the values such as transparency fairness and accountability in AI development and deployment in a world world where ai's capabilities are expanding rapidly it is crucial to remember it's not just about the outputs of AI but the outcomes for society let's ensure that as we navigate this era of innovation we prioritize ethical conversations and responsible regulations because in the end the impact of AI on our lives and our democracy can just not on what it can do but on what we choose to d
o with it let's shape a future where AI enriches not deceives what empowers not manipulates the future of AI is in our hands let's make it a future that we can trust focusing on outcomes not just outputs thank you by the way this is just a deep fake image that we will all recognize that moment in our history already where someone put that image up and within literally a half hour the um stock market went down I think it was a half trillion dollars something like that lots of money um until it wa
s clarified that yes the Pentagon was not uh um um blown up um that shows you just how quickly these Technologies can affect our society so it's my pleasure now to bring up our first uh legislator who's a fantastic consumer Advocate um uh Senator Padilla um is the author of Senate Bill 1228 which will bring up our little um poster board on he's a lifelong public servant has had many great uh uh Public Service stints as well as in law enforcement um focusing to protect our kids which is of course
so critically important um stin is chair of the um California Coastal commission to protect our coast and uh he just has really been fighting to protect consumers and we know we're all consumers of political information right now so thank you for coming up and talking a bit about your bill Senator thank you thank you well good morning and it's a pleasure to be here uh with the initiative and with my distinguished colleagues in the assembly uh to talk about the bill package and the collaboration
but first and foremost you know I've always said information and knowledge is power and intentionally manipulated what passes for information that is indistinguishable from truth is that threat to popular sovereignty and no less an invitation to tyranny and we are living in such a time where younger and newer Generations that follow many of us are literally unable ill equipped and unsupported in their inability to distinguish reality from manipulated information and a perception of reality leav
ing them and our society and in a republic our electorate more and more vulnerable to manipulation I'm proud to stand here with cited we're entering an election season I'll reiterate it was mentioned a few minutes ago uh in which disinformation as it was said a few minutes ago powered by generative Ai and foreign governments can pollute our information ecosystems in a way unprecedented in American history Bad actors and foreign Bots now have the ability to create fake videos and images and sprea
d misinformation to Millions at the touch of a button users today have no effective way of distinguishing content produced by a reputable news source a trustworthy person or a foreign bot other Industries familiar and necessary banks credit card companies are required to take steps to verify consumer identities and to understand their activities to protect against fraud corruption and illegal activity online platforms such as social media or video sharing websites have no such requirement at the
moment no other industry is allowed to self-regulate like this so why do we make an exception for the omnipotent Tech Industries we're still learning of the incredible harm and potential harm that social media has had on young people we know of Russian attempts to be specific to influence our election Integrity here at home we need to ensure that our content platforms protect against the kind of malicious interference that we now know is very possible this is why in partnership with cited I hav
e introduced Senate Bill 1228 which would require large online platforms to verify the identity of users with more than 25,000 followers or that have shared more than 1,000 pieces of AI generated content the platforms would be required to verify the name of the account owner their phone number and their email for accounts over 100,000 followers are sharing more than 5,000 pieces of AI generated content the platform would be required to verify through government issued identification if the accou
nt is unable to or unwilling to verify their identity the platform will be required to flag it to alert the public and users who will then be able to opt out of receiving posts from these unverified accounts and sources protection like those articulated in the bill and others um are necessary and required to not relive the hard lessons we learned with the Advent of the worldwide net these are measures on how California can continue to be a leader on Election security and look and I look forward
to working with my distinguished colleagues was cided and helping to persuade our colleagues in the legislature to move these important pieces of legislation forward and I thank you okay um obviously these are ambitious challenging and critically important and our our next uh speaker is uh a true leader uh for many many years on Election issues in our democracy um assemblyman uh assemblywoman pin um um has uh I look at her wealth of experience on elections Administration and she served for over
27 years as the chief elections official in Santa Cruz County she currently chairs the assembly committee on elections um and is spearheading a number of bills ensuring accuracy and protection for California's democracy including AB 2839 and I can just tell you she is just um so insightful about these issues and we're so grateful to be partnering together we're really grateful Gail thank you come on up thank you drew thank you all for being here thank you to my team and uh my incredible colleagu
es who are in this with me um my name is Gail Pellerin and as Dre said I served as the chief elections official in Santa Cruz County for over 27 years I'm very familiar with all the different ways people try to impact an election um and AI is that concerns me very very much and in my role now as a State Assembly member and chair of the elections committee uh it is my um dedic you know I'm dedicated to protecting democracy and making sure our laws um M um ensure fair and transparent and accurate
elections uh there have always been bad actors who try to wreak havoc on elections I remember an election where people distributed door hangers uh which gave people the wrong polling location in order to disrupt the election outcome but now the widespread access to generative AI is making that type of damage so much more sophisticated and easier to disseminate this year California is entering its first ever election during which disinformation powered by generative AI will pollute our informatio
n ecosystems like never before and millions of Voters will not know what images audio or video they can trust some of you may have heard about a robocall of President Biden we heard about that earlier encouraging voters not to participate in the primary election in New Hampshire a few months ago that AI generated robocall was a prime example of Nefarious uses of AI which seek to disrupt our elections imagine another example in which a deep fake video depicts a candidate running for office office
doing or saying something that they did not say or do or a deceptive video of an elections official saying that the rigged uh they rigged the votes in favor of one candidate with generative AI deceptive videos like these can be more easily created and distributed online this type of Technology can create insurmountable damage to election outcomes that's why I'm very proud to be working with cided on AB 2839 to prohibit Bad actors from Distributing AI generated mailers rooc calls and video ads t
hat are materially deceptive to voters the bill prioritizes a rapid response to injunctive relief and applies to dis disinformation about candidates running for office election officials voting procedures and other matters surrounding the Integrity of our elections getting the right information voters is crucial to a functioning democracy and it becomes very difficult to ensure the Integrity of the election when these deep fakes are spread online by phone or through the mail to intentionally mis
inform the public I look forward to working with sided and my colleagues in the legislature in passing this important Bill thank you so [Applause] much the third member of our incredible legislative team has a long history in fact is really the leader of of dealing with deep fakes in the state of California five years ago um assemblyman Burman was preent to say this is a danger to our democracy we need to deal with it and he managed to navigate through a challenging legislative environment to ge
t the law in place that we're now going to to strengthen with his help um particularly uh impressive to me is that assemblyman Burman represents the heart of Silicon Valley and um needless to say has to balance these competing interests of innovation while at the same time trying our best to protect against U massive disinformation that these platforms can uh lead to um um he previously chaired the elections committee so uh is is very well versed in these election challenges as well of of course
um and is now the chair of the business and professions committee and we're very excited to be partnering uh with Mark on this bill AB 2655 thank you thank you s women well thanks so much Drew and thanks to the team that cited i i i this is going to sound dorky I got really excited um when I when I first heard uh that your organization was created and and the work that you were focused on and and I'm really grateful to be teaming up with you teaming up with my fantastic colleagues um from from
the the Senate and the assembly uh to focus on this really critically important issue and and as Drew mentioned five years ago uh I authored what what appears to have been the first deep fake bill in in the nation I did this after watching a different video uh which was a video that from Jordan Peele uh the actor and director and producer who did a deep fake on President Obama back in 2018 uh it was funny I guess former President Obama if it was in 2018 it was funny but it also set off alarm Bel
ls uh as you could easily see how this promising technology could be abused by Bad actors to undermine our democracy and destabilize the country the potential harm was fairly limited back then however as the technology to make a deep fake was expensive and not widely available just a few short years later the technology is better cheaper and is now accessible to anyone with a smartphone today it is far easier to make a realistic deep fake and we are seeing AI being used to undermine elections ac
ross the globe and here in America as technology changes so too must our laws it's not just policy makers that are concerned the California public is demanding that we act to minimize disinformation and protect our democracy according to a recent Berkeley IG poll 84% of California voters overall are concerned about the dangers that disinformation deep fakes and artificial intelligence pose to this year's election in 2024 nearly three and four voters 73% believe that state government has a respon
sibility to take action to protect them from this deceptive content and this isn't a partisan issue a majority of Democrats Republicans and no party preference voters believe this way that's why I'm so proud to partner uh with cided on ab26 655 which will ensure that online platforms restrict the spread of election related deceptive deep fakes meant to prevent voters from voting or to deceive them based on fraudulent content uh and this is just you know all of our bills combined will create the
regulatory framework that's really necessary to protect our voters uh to to protect our democracy um and and uh you know that's why I'm especially uh you know proud to be a part of this very comprehensive effort to make sure that we as as state government address uh while still fostering The Innovation that this technology can uh and the potential and the promise that this technology will will uh you know can create for society also minimize the risks and the harms that we know it can also creat
e so uh really proud to author AB 2655 I'll turn it back over to Drew great and I think what we'll do is we'll have you both come up as well and see if we if you could stay something and we'll see if we have any questions most of the Press people these days talking about Silicon Valley and Technology um most of them now get to do their press work through live stream oh perfect yeah so they get to sit at home in their pajamas and uh and not all of them though but not all of them you could wear pa
jamas but um we're grateful that you're here got her phone record um so I just want to check to see um K if we have any questions that have come in uh from folks on the live stream yes we have okay first question is what happens if the legislature does not act now on this issue okay um you want to take that sure go ahead I mean I I mean I think we're seeing what happens if the if the legislature doesn't act now which is we're seeing Bad actors whether they be foreign Bad actors domestic Bad acto
rs uh utilize this technology utilize AI uh utilize deep fakes to influence uh to to create the appearance appearance of candidates or elected officials doing or saying something that they did not say or do um and and uh that influencing or have definitely having the potential to influence our elections and so um you know we're seeing it happen in other countries we're seeing it happen in the presidential primary um we I think with pretty good certainty can know that it's going to happen in our
elections in California which is why it's so important for all these bills to to get pass and sign into law well and I would just please I would just add that uh my colleague assembly member Burman was uh on the Forefront of this in a Trailblazer on this issue back in 2017 I think 19 okay but we also have the office of cyber security with the Secretary of State's office that was 2017 right 2017 or 18 yeah so so so we have been working in this space and will continue to do so to stay ahead of the
uh Bad actors absolutely I have a question if you'll take an in person um only if you're in your pajamas um so are we aware have you guys worked with any other like lawmakers or officials in like the EU or the UK or aware of like similar efforts in other parts of the country or the world in respect to a broader effort that I'm undertaking in the Senate uh and and you'll be seeing a little bit more about this in the coming weeks in terms of the scope and impact of AI generally uh uh uh I think u
h you know the European Union adopted a set of principles that are focused more on consumer protection rather than on proliferation Innovation uh we saw the White House Office uh recently issue a set of guidelines and and everyone from the NGO sector to the US Chamber of Commerce have issued pillars or principles that they think should guide so the short answer is I have no doubt that my colleagues and and certainly my office has engaged diligently with every level of government and with thinkin
g uh practitioners people in the industry and people uh looking to protect against consumer exploitation um around the globe nationally and here in in California uh I think it's important to also remember the urgency to act is made even more important because to be blunt the dysfunction of our own politics we have divided government at a at the very same time uh and we we have you know 50 sovereign states and other territories that have a a position to play here at a time when um this is being d
eliberately introduced and manipulated uh and the government can't seem to get on the same page I hope that California as we lead in so many other ways around the nation that this can be an example that the largest state in the Union uh can act together I would only add uh with these members that um a very basic function that cited is playing is interfacing with uh folks from the EU one of our members right now is actually in Brussels meeting with EU Regulators we're wanting to bring that inform
ation to these legislative leaders to make sure that we are on exactly the same page as the dramatic work that the EU is doing I point out to people that the EU is not constrained by one of the most important um um Bill of Rights we have in this country called the First Amendment and that constraint that we face really makes it difficult to craft the bills that are so important in the state legislature California will certainly be a leader on these issues we have more bills in the hopper right n
ow on AI and election protection yes than any state in the country and we've led the way on privacy obviously that other states have followed and I have no doubt that with their leadership we're going to lead the way on these AI protections as well just briefly also you'll see um at least in in the text of my bill that I take a a relatively similar approach that the EU did of a tiered approach uh so identifying what are the highest risks um and having different um you know different different so
rt of obligations necessary depending on that level of risk which I know the EU took as well with deciding to prohibit certain things and then also regulate uh other types of of content so definitely uh you know a lot of a lot of collaboration another question Ken yeah there's a question around Senator faia your bill um what are some of the First Amendment kind of implications to kind of requiring you know large accounts uh for having to verify the information I I would argue that the implicatio
ns for the regulations suggested in my bill go to um the Integrity of what is being profer as information uh it isn't uh it isn't something that's going to stifle free expression it is only going to uh inform the consumer of information so that they can make an informed decision uh and put that information in proper context absolutely excuse me sir we're doing medical chairs here yeah sure so folks have talked about kind of the efforts in California so why California specifically there parallel
efforts in other states maybe some contrast with other states and also why California is important why should California lead the way well California has always been a leader uh and especially in the area protecting democracy so uh we were the first state to implement provisional voting back in the day when people were turned away from voting on Election Day because something was off with their voter registration and and when we started that other states um came on board as well so I I think Cal
ifornia is absolutely the right place to start with [Music] this I I'll just also add because Congress isn't and and and and so and so we need to uh and we need to do what we can to protect our 39 million residents and and we need a hope as we've seen in a lot of other uh policy areas that what we do gets adopted by other states across the country and and so in regards to my bill this is actually building upon what what I did five years ago but it's also because we've seen other states improve u
pon what we did five years ago um and so you know it's constantly trying to iterate constantly trying to improve um as as the landscape as the technology landscape shifts and as we identify new harms excellent well I think we'll stop there because you have to actually work yes thank you all for coming obviously the hard work now begins this is the easy part to say what we want to do we're hoping that this turns out to be a bipartisan effort and I say bipartisan because these issues are affecting
everyone right and and if it can become bipartisan and if there is sufficient support some of these bills just might be able to be urgency measures which would mean they would take effect immediately upon Governor Nome signature and we hope that that is the level of success that we'll be able to have in this process to protect everyone so thank you so much for coming today all right app you thank it fun going be good apprciate your leadership thank you
Comments
Last thing I will ever trust is CA political views.