court -- >> navarro's lawyer -- >> set up a road map for all future constitutions. as i said -- in the courtroom today, i am not only just the first ever with charges but i'll be the last person because of the road map we have established. [whistle blowing] c-span now is a free mobile app with unfiltered view of what's happening in what live an on demand keep up with the days biggest events with live streams of floor proceedings and hearings from the u.s. congress. white house events, the court, campaigns, and more from the world of politics. all at your fingertips, you can also stay current with the latest episodes of washington journal and scheduling information for c-span tv next and c-span radio plus a variety of compelling podcast c-span now available at the apple store and google play scan qr code to

download today and visit c-span.org/now. c-span now your front row seat to washington any time, anywhere. ♪ ♪ welcome back to washington journal, we're joined nowy steve he's new york university and the executive director of the center for social media and politics. welcome to the program. >> thank you for having me. >> for talking about artificial intelligence deep fakes can you first explain what that is. >> yeah. so pretty simply a deep fake is manipulation or fabrication of an image audio recording or video through automated means with the factth of making something sort of appear to happen that didn't actually happen. and fakes have been discussed of potential threats in the last two major election cycles in 2020 and 2022. so i want to quickly explain why this moment is different. 'v had technology

manipulate content in different mediums and know that hollywood movies have been doing this for many, many years but expensive and time consuming it took train specialistmi with expensive software and required many, many hours. now, with a.i. we have a different momen where the technology to create deep fix sort of is demock are and lowerd access. >> so let's take a look, this was a -- fake robo call of what sounded like president biden - calling new hampshire voters ahead of the -- the primary telling them not to vote. take a look. >> a bunch of ma malarkey we knw value of voting democratic when votes count it is important to save the vote for the november election. need your help in electing democrats up and down the ticket ppg voting this tuesday only enables republicans in their top

again. your vote makes a difference in november not just tuesday. again that was not president biden although sound an awful lot like him your comments on that steve as far as -- being able to tell the difference. >> it did sound a lot like him. if i had been on other end that have phone call i couldn't tell the difference it wasn't him. i think this is a challenging moment of the information environment and i think what sort ofe interesting that we've seen when it comes to deep fake is a lot of deep takes we've seen break through, have been autoowe only they haven't been video or images. i think one of the reasons is because audio is very hard to fact check if they want a fabricated video of joe biden his every move is watch and recorded and you know while -- there might be a few minutes or few hours before it becomes fact check it is fact checkable. this is much harder i would say,

you know, it's a challenge moving into 2024 with that audio clip for example it would have been hard to make using technology you can jap load minutes of a politician voice and create a a.i. version of them saying whatever you want. so you know, i think that -- you know, the public should go out with a healthy bit of skepticism to get back to this later not too much skepticism. >> if you would like to join our conversation about a.i. and its use in campaign 2024, you can do so on our lines bipartisan democrat 202-248,000 republicans and independents 202-24 000 joins on text and social media democracy but how easy is it to

do now could n i do it? what kind of technical background do you need? >> i think that you probably could do it and i don't know what your technical background is. you know, that i think these -- technologies are very accessible. depending on the different company, different companies have different rules around what their technology can be used for. so certain companies like open a.i. for example would express product this type of content from being made using, using any of their products. but there are a number of products out there many of them are very, very good i would say they're not quite sort of as easy as just you know -- you know click and play. but they're getting there. >> and -- another part of this steve could be what somet people are callig the liars dividend. can you explain that? >> yeah. so the liars dividend what that refers to is that -- information environment where

there's a lot of anxiety around false or manipulated or fabricated content circulating and allow actors to do is claim that something that did happen that actually did happen -- that it didn't happen so you could imagine sort of in the context of 2016 -- that act of hollywood tape coming out -- which again is sort of an audio -- clip without video or a photo attached to it as we just saw with the joe biden clip it is very easy to make this sort of audio content you can imagine donald trump having claimed that it was false. so -- you know that is what the liar dividend is and it will give sort of strategic actors the general context in which to claim something that's true is actually false. >> let's take look at a couple of other examples. this first one is an ad put out by the republican national committee just after president biden announces reelection -- campaign last year, and it features a.i. generated fake

images. >> this just in, we can now call the 2024 presidential race for joe biden -- ♪ ♪ >> this morning invadings taiwan. >> financial markets are in free fall 500 regional banks shut their doors. overrun by a surge of 80,000 illegals yesterday evening. >> close the city of san francisco this morning. i need escalating crime and fentanyl crisis -- >> who is in charge. it feels like the train is coming off the tracks. >> and these -- i mean, that's using a.i. generated images of the future -- do you think that should be allowed? >> look, so -- both at the federal level and state level -- you know, both policy, you know, lawmakers and regulators are considering what to do here. i think i want to -- point to colleague that you've

seen brennan and they argue that -- sort of two things that first we should really public policy should target the electoral harm being done not the technology that are being used. as they mention it did sort of access to the technologies, but really well finance, you know, national campaign. and how to access other sort of more technologies in order to create fabricated images in previous years if they wanted to so should really focus on what the electoral harms that we're nervous about rather than underlying technologies given that at this point you know, sort of -- these out of the bag and second is we're -- still sort of early days here. so a lot of policy rather than sort of perhaps jump into guns and focusing on wrong things should really focus on promoting sort of learning or new knowledge about these -- how generative elections may have political lives so you know should be aan big focus on

transparency so making sure that any time, you know, generative i. viewers at home would know it. but also promoting research to really understand what the effects off this sort of content is on voters. >> when you say transparency, i mean, if you've got bad actors out there they're not going to be transparent that they're using deep fakes. >> yes not going to be transparent if they're using deep fakes so it might put some of the -- on, you know, the sort of mechanism by which it's reaching whether broadcast networks or, you know, television -- requiring them to due diligence in ordero to apply labels, socl media companies applying labels. or, youme know, i mean, i know t the state level they've been considering there being -- , you know, certain penalties for not disclosing. >> you wrote a piece for brookings, steve, that

essentially says a.i. cub misunderstood and fear can be overhyped. explainar that. >> so i think a lot of our fear moving into 2024 is specifically around sort ofnt relationship between a.i. and misinformation. so what we do is we start with a review of the academic literature on, you know, what do we know in the last ten years about the way that misinformation has or hasn't directly impacted elections and by directly i mean change people votes. and -- you know, we sort of me the argument reviewing literature that actually we've been potentially too concerned about the impacted of misinformation. that you know, news doesn't really make up majority of people media consumption when it does people tend to consume high quality media., and that misinformation when it does exist tends to be public

concentrated in the extremes either party -- but that, you know, does tend to reach most americans. and so for us -- rather than saying you know, using that evidence to argue that a.i. devil won't impact 2024 wens face what might be different about this moment we're in and we focus on three different ways of a.i. -- you know, might make misinformation impacted or directly impact elections more -- the first is to make it more pervasive there could be more of it and there's a simple math which, you know -- misinformation needs to break into the mainstream if there's more of it there's a higher probability that it does -- the second is that it could make it more persuasive or more effective that it could be better or targeted -- you know, there's reports in 2016 of the harms there -- you know, tweet riddled with grammatical areasing and finally we can have this moment where

tiktok is this different social media platform which really sends content to users who haven't opted into a particular social networkte that's not thai have to follow you you need to get your content but instead it can reach f me and that dynamicf information defusion might change, you know, the effected ofof who or change who misinformation is reaching and that's what effective has. >> we'll get to your call shortly the numbers are democrats 202-248-8,000 and independence -- you can start calling in now -- public opinion about this. the a.p. did a poll in october, of this past year, and i'll read you some of the results and■i he you commentren it so the question is -- how concerned are people about the a spread and increase o fale information? 58% said they were concerned.

54% have not read o heard much at allbo a.i.. and m 30% report they have usedn a.i. chat bot or an image nese numbers? >> yeah. so i want to focus on the first number which is the level of anxiety here. you know, i think we've talked about in this sort of right after the joe biden stuff there should be a healthy level of skepticism. and by healthy i mean -- people should be sort of -- understand that h these technologies exist as they have certain capacities to produce the type of content that you've already played on the show. think, you know, here as social media in politics toipght put out there we are nonpartisan research center so what we care mostly we care about democracy and sort of democratic society and there's such a great anxiety around use of these technologies to create sort of false or fabricated

content that people that -- youo know, that trust in the general information environment continues to erode. and we're still seeing that people tend to consume sort of high quality credible content some reputable sources like c pan so i think that -- going into 2024, you know, a little bit of, you know, skepticism of understanding that these technologies exist is important. butf not so muc that people stp trusting what they see. ening debra is next a republican in westchester, ohio, hi, debra. good morning and thank thank yor taking my call and we'll put a position that we have to always question original source. when you look at section 230, i'm thinking by this time we should have solved and resolved that and we'll never know what social media platforms would

have been had we held them accountable. other concern that i have is privilege speech as described in our constitution. so who -- you know, we present speech when it is generated by a human. what about speech that is generated by -- chat?ra these are major constitutional questions, and we're not real good at resolving some of these as we're going going to be developing and discussing these issues we wait until it is too late and everybody is used to these various things. but there has to be something some kind ofri a disclosure on every political ad that says there was or there was not artificial intelligence in generating the images. >> steve what do you think? >> yeah so, i mean, i think that -- sometimes transparency is sometimes the greatest disinfectant here and

transparency should be a big piece of this especially with politicall ads. i want to pick up on something that was mentioned around social media.a. that i read we sort of, you know, when it came to the development of social media and how quickly it is gong to change society i think we were slow on a nonlawyer never to comment on that but instead comment on the research side. i think that -- when it came to giving researchers like us -- the typee of data and, you know, importantly funding that we needed to do this type of research -- it was too slow so we're still catching up on really understanding effects of social media platforms so here when it comes to development of a.i. technologies pace of development is dizzying i hope we don't make the same mistake again so we have this society approach to understand what the effect of these technologies are. larger so that lawyerings and policymakers as they go out and

either -- you know, argue in courts or halls of congress about what to do here. that they're given really high quality information in order to -- noshed to do their work. >> ira in lewisburg, north carolina, democrat. >> good morning to you. are you hearing okay? >> yep go right ahead. >> okay. as attorney -- i can tell you that law enforcement agencies are already using the technology that you're speaking about. on your clip -- initially that says one thing and if you don't make a copy or youu don't video, video -- week that video would have changed you will get a video -- but it will say something different audio will say something different that's already happening. it's if you lose law enforcement

only time they have for anything they do -- is when someone takes video or audio. if you have not video or audio clip they're not going to -- back -- things like that right now. every one of my video clips any one of my audio clips have been altered and file something with the department of justice to have this case -- >> all right. >> your comment. >> yeah. so this is, you know, in terms of -- my director research you know i tend to focus on elections and -- certain electoral politics but i will say that, you know, we are seeing the sorts of harms that occur elsewhere and law enforcement context is one of the primary ones that just sort of me as a citizen are worried about and if i understood correctly, it related to a liar dividend that i mentioned that

suddenly gives the general context for a police officer who potentially said or did something, in fact, to claim that that video was fabricated using tools. ... he said it does this technology get dark money from overseas? do you know anything about that? >> yes. the challenge right now is there are many technologies. i wish i'd said this earlier it's not a monolith here.

and the actual technology itself what models are being developed toward what end but a also the companies. i think right now there are companies acting and quite gd to consider the potential harms been dealt with open ai which just last week release or irelection policy that i found o be quite considerate and hope they can continue in that regard. there are other companies her in the u.s. and abroad that are not putting up guardrails in terms of the actual funding sources i am not sure. one of the places i hope we do focus is to hold these companies to high standard of the policy to be had on the books as well as being force in the policy. cracks next in hendersonville, north carolina independent. >> caller: hello. i am currently setting artificial intelligence programming and college. one of those things we recently

discussed and how widely said it is paired anyone has a smart phone uses ai. as invaded every aspect of our lives when it comes to technology ai has been there for a while. with the fear of what it can do what people can do using it for elections can cause a decent amount of fear and i do not want another ai winter to develop. as i'm trying to get a degree and make a career out of it. your comments?

when the challenges we talk about ais what do we need? it's quite a broad term. one at the jokes and can speed or science anything that doesn't work yet. one of things we are concerned about our fear of the effects will end up being greater than the actual effects. so we focus on what democratic consequences that might have on people become too skeptical of the information they see we see general erosion and trust in the information environment. it does have potential economic impact of the fear was special special andit comes a globalci . >> or text from jimbo a bakersfield, california. the fake ai generated robo call of present invited uc new

hampshire is an omen of the your head? and i guess the new hampshire investigating up it will have to find out as far as the judicial side of it. do you think more groups will be encouraged to this kind of thing? i think it is an overlook of their head. what i mentioned especially when it comes to audio that is going to be quite more challenging there's so little other contacts for people to use in order to potentially fact checking themselves. we've also seen over the past couple of years the number of robocalls has continued to expand. that could be a place where we see a lot of concern college focus on stands are especially concerned about automated robocalls. when it comes to election i think were going to see strategic actors are motivated

actors both here and abroad tried to use the technology to shift the information landscape. in his sake and all of society approach. i think lawmakers should definitely understand the gravity of the situation but also make sure they are leveraging the high quality information to pass really good allows. but journals to recover the space researchers to study it and private companies to be really good actors. and to try to understand elections are very important democratic processes and do their part. bound, independent good morning. >> caller: good morning, thank you for taking my call. while technology is one thing, it is a wonderful and we all have to make an effort to learn as much as we can and understand it, besides technology there is

something called moral and ethics. we have as a society today it is the accepted thing to do. the majority of the population if it's accepted for them who's telling the truth? thank you veryh. much. >> ask any comment there? >> yes. they is to be public accountability for using these technologies in harmful ways. what other things i wanted to mention it as we previously were talk about transparency efforts in this regard. one ofhe the colors mention the content there should be laws passed to make sure it's artificial intelligence is used it is- labeled. one of the pieces we really focus on here when it comes to

truth is the devil is in the details. it matters whether the label comes before the adder after the ad whether it's in a large text or small text. especially as lawmakers they focus on the little pieces that impact the public. >> john is in warrenton, morning. republican good >> caller: good morning i like that money question about dark money coming in from overseas. i almost see this video that you show as the initial stages of the computer age are we played pong. ai will be as are much more complicated interesting things. so for example i get 25 robocalls a day. they have changed my phone i cannot answer calls to my phone anymore. i have to say i don't really

think they are trying to sell mo something. i think it's an attack appear to think of some sort of weird way to change my life. remember how much in the '90s on the internet came on how much it changed our life and push the economy for their so many new businesses. i wonder if ai is going to the same thing. it's going to build to solve a lot of problems and one more questionn, i can't remember ver. >> i have a question for you how you think ai could impact the election, when you think of that? yes. i am early for c-span count i was listening the first day c-span was on radio back in the '90s. and i was what they called a news junkie in the '90s and college at you mbc.

how's it going to affect the election? i would have to say -- that little o bit of extra line they could do with ai is nominal. that's my opinion. we already don't believe what's going on. let's talk to zion and germantown, maryland good morning. the last go around the governme i not able to make a substantial change in the average american's life. with this, we are buried even videos on our phone may be an accident may be a protest rally, may be bigotry. those against the opposition when it comes to thehe companies ownedt by one major conglomerate

and they failed to consider bottom-line they could get into our phones and alter these things. their biggest challenge is having been that important in a recent and technology. they let things go by like recently with apple finding their change in the product to be manufactured in india we did donot hear about stringent restrictions. i don't have faith the government will do something. have they developed a new agency?

we a create a space for us. we got your impact -- how impactful the government will be. >> one of the glimmers of hope in the government internationally trying to do itn mention they did not do during the social media era the beginning of social media is they are trying to understand the trade-offs between democratic benefits and democratic harm. it's getting good research the laws passed are high quality. announced a large new funding program in ai to do exactly this.

governments move much faster. whether it's joe biden's executive order for the bill at the state level where singh or d i think again they are learning from some of the mistakes are social mediaoc era and i think n be quite impactful for. >> just mentioned the executive order. i have ited here on the fact sht present buying orders trustworthy intelligence. that came out in october. what impact do you think that's going to have? facts in terms of the impact is going to have it's a little bit too early to tell. i'm going to be needed to be updated the technology of a certain size and i think as technologyas advances we are gog to need to rethink that at one

point may have focus on large computers and now computers are quite small and handheld. in general a gesture in policymaking towards trying to move quickly, trying to learn much more quickly about technology and we have seen federal government do so in the past.ve they brought experts into federal government to understanding technology deeply. they have engaged with everyoneh from leaders of these companies to academics, to also which has been really promising, civil society organizations and community groups acting with the folks were ultimately going to be impacted by the technologies to make sure they're mitigated it's a little bit from early days to understand or pinpoint the impact ofder but i think inl it's really exciting to see the general orientation policymaking towards working in the past regardless go to augusta,

georgia next independent,. >> good morning. >> good morning and thank you for c-span. we enjoyed watching in particular on this subject. i would ask your guest, when the government says you have to label ai, how quickly they may or may not have ai in their material? that is the problem. one step ahead of you forever thank you. absolutely. when it comes to labeling we have seen the timing is a big tpiece of it people tend to be exposed hours or days later it's not going to have the impact.

i think again the devil is in the details of policymaking and the mechanisms are all there. and your knowledge be to do numerous on this call another area have positive effects as to do identification and labeling work. again it's early days i agree with the call at the time will be a really big piece the nefarious actors all want tote sidestep to get ahead of them. >> john is a republican in >> thank you so much for taking my call. being the technology was created all the social media and all of the communications is controlled by the government.

not not hurting children all over the world therefore dictator governments that we work with being computers this gentleman said the social media was more complicated than this other ai but other countries are building robots. to replace human beings. also law enforcement who's been killing the majority of people ofty color basically all over te planet. >> john we are close on ai and the campaign. jeff something specifically about the election? >> yes. i think the whole system is rigged. it's been related to the king of england were getting a little off the subject was talkbj to

jack. good morning. i am completely for technology. and artificial intelligence. but it seems like the country itself, what is evil is good. and what is good is evil. i have the impact on the statement i just made. the republican side or the democratic side are just individuals who own operations and whatnot. and what your thoughts on that? and thank you. >> we also have this from michael on x who is going to decide what the reputable forces are?

what the general statements around good and evil. we have the generalized fears of technology either we see these overly or do morris and ruining everything. very specific benefits. regulars do their job better.taa richard more nuanced way. in terms of credible sources we have thought about journalism and media over the last 25 years. i am not going to -- again we

are a nonpartisan researcher center. over this media company or that media company and said what i focus on when it comes to credibility is whether an immediate company as a process in place that you would consider to be credible? whether they fact check their stories whether they retract stories they are shown to be false. whether they do things like disclose who writers are, who their funding sources are pretty think credibility lasts as an outcome. people could not have diverse political perspectives and ideology. but a commitment to a process that can come from a variety of sources with different >> -park illinois good morning. >> given what happened in the

2020 election and to be coordinating the former twitter which is now x in a nefarious way. they are supposed to do something new technology good and to help the humanity progress. but, given what happened in the 2020 election by her very own government it worries me at may be designed to be used for nefarious purposes against humanity as opposed to helpingan humanity into control humanity. that is what worries me because folks designingho ai are compans like google which is on the wrong direction. as well as meta.

they should be checked who they are coordinating with, who they are hiring to design and programs as election interference in our government and biden with the administration there also coordinating we want our technology and very nefarious ways to control our elections and soan forth. so that should be investigated from the angle of this administration that statements shall be as ai ahead of everybody in order to control the.ho >> okay good steve come in, there? >> my comment is -- by general comment here is one of things we hear when we go out and talk to folks and solely from the public opinion polling that you mentioned before with the general anxiety around the newl technology. researchers need to move quite

quickly so we can bring high-quality evidence to the public i don't think there's any evidence right now that ai is being used in the election context for controlling the public. but what i do here at the general anxiety and i hope we can get high-quality information out to the public quickly. hookser montgomery village, maryland independent too. >> hi. thank you for putting me through. i was just going to say i heard the woman talking it's not necessarily the technology that is the issue it's the people were going to be in charge of it. mayor is like oh no the technologies going to take over and will have to be at the >> and call because at the end of the day it's people writing the algorithms there's a lot of corruption in the government i'm not surprisedve people are hesitant about it you understand human nature. will have the capacity to do good and bad. it's a matter of choice.

that's all i had to say. focusing on the outcomes and give evidence of this is good or bad. you can manipulate what you"- wt but the human nature issue and ultimately my final point is at the end of thete day human probm seen human solutions. hewitt solutions. technology is just a tool. i think that's a little bit deeper than that but that is all had to say. thank you very. >> last words? as a social scientist that the last thing the calleren said is speaking my language. completely agree. there are humans that need human solutions and we really need toe understand and didn't intentionally technology condite technology themselves i was happy to see governments were capacity internally print dosage approaches from a human center

to understand how the general public is impacted by this. it's early days. i don't do have a full understanding of how it impacted when it comes to election we can better understand how to mitigate the harms we talk about here but this is going to impact people in every a way. whether their work and labor whether it social lives i think we really need to approach this from a human point of view and i hope people at home that's the way they are thinking about it and journalists and government are really making sure people are the center of the way there thinking care. >> university center executive director bird we can findhem online. steve, thank you for joining us today. quick things have me. summer trade advisor to former

trumpets for months in prison for defying a congressional subpoena and relation to the house investigation. that's into the generate 60 west capitol attack political rights the sentencing comes after nearly two years of proceedings that became entitled a couple kind of questions about executive privilege and whether his defiance could be considered criminal he believed trump had won him to protect the confidentiality of their conversation. the publication goes on to say the judge declined to immediately rule on the bid to remain free while he appeals his conviction. the ultimately rejects that he would likely be the first senior trump white house aide to spend time in prison. and his lawyer address reporters outside the courthouse. here's some of what they had to say. >> what is important to me sir that it has always been important to me is the underlying constitutional issue associated with that. i way to say that my standing up

we have already achieved an important victory. the department of justice attorneys after pressure not just for my defense team but from the judge himself finally acknowledged senior presidential advisers such as myself cannot be compelled to testify. we have acknowledged they would not prosecute in that case. there is very much a constitutional disconnect here. an important mission is to settle good law in an area where there's nothing. remember, please, i am the first person ever charged with this crime in the history of the first senior white house ever charged. everett charge. and, and what my defense team

has a beautifully done is in cooperation with the court. has set up a roadmap as i said in the courtroom today i'm not one of the first minutes charge i will be that last because of the roadmap we have established. >> and set a predent vital be speaking at the south carolina democratic party first in the nation dinner in columbia. watch the president's speech alive at 7:30 eastern on c-span or free mobile videopps is been now or online at c-span.org. ♪ discovered the heartbeat of democracy it was c-span's voices 2024 as we engage voters nationwide asking what issue is most important to you in the selection of why? >> the most important issue to me as a healthy congress because

nothing is getting passed, the debt ceiling keeps getting postponed because no one can agree on anything. so i would like a candidate that can bring everyone together. >> my name and the election is very important to me because issue on the economy is in shambles and i hope we have some and the white house. we are seeing an unprecedented amount we need a strong president who can tackle the border crisis or. >> for. >> hello i am tiffany i believe the most important issue at this time is our democracy. saving our democracy's upmost importance on ancestors in so many americans have fought for the right for everyone to vote. this is the most important issue of our time. on the most important thing we

need to focus on. >> he spans be a part of the conversation. ♪ ♪ ♪ c-span is your unfiltered view of government we are funded by these television companies and more including cox. ♪ is extremely rare. but friends don't have to be. when you are connected you are not alone. ♪ c support c-span as a public service along with these other television providers. giving you a front row seat to democracy. >> a senate hearing on the reauthorization of the national flood insurance program which is administered by fema. the hearing comes as congress works to renew the program which allows for the sale of flood insurance which is not included most standard insurance plans. from that banking h