How US Campaigns Are Already Using AI to Try to Sway Voters (msn.com) 61
39 U.S. congressional candidates "reported paying for an OpenAI subscription this election cycle," writes the Washington Post, citing campaign finance disclosures.
This means the campaigns "are using the technology to court voters — despite limits imposed by leading AI companies to protect elections from the technology's risks." Two [candidates] said explicitly in filings that they had used the subscription for advertising, even though the company's policies prohibit candidates from using their tools to generate ads. Another candidate disclosed using AI to draft and personalize political messages or create synthetic media, though they did not specify which software they were using... Around 30 political action committees and parties have reported payments to OpenAI, with the Republican National Committee ranking as the company's largest political spender at roughly $9,700, the analysis found... Political consultants say the disclosures understate how many candidates are using ChatGPT and other AI tools to craft messages for voters... "For the most part, people are using it to write their emails, write their ad copy, write their scripts," [according to Eric Wilson, a Republican digital strategist who has advised campaigns on AI]. "But no one is going to go around saying, 'I'm using AI.'"
The lack of transparency from campaigns reflects a paradox facing politicians: Generative AI is growing ubiquitous, and candidates could be at a disadvantage if they're not using the tools. But voters are less likely to trust messages they know were generated with AI, studies have found, making campaigns loath to disclose using it. Political consultants expect that as Election Day approaches, more campaigns will outsource AI-generated ads and materials to super PACs, much as they do now with negative ads. Katie Harbath, CEO of the tech policy consulting firm Anchor Change and a former Meta executive, said there are many parallels between negative campaign ads and AI: Voters say they find such ads distasteful, but politicians keep using them because they work. "Typically, you would give some more negative stuff and more risky stuff to those [outside] entities," said Harbath, author of the upcoming book "Disrupting Politics: A Front Row Seat to the Collision of Technology and Democracy".... "You're starting to see the tension on the left about using it, where they're saying, 'If the right is using it, why aren't we?" Harbath said. "It can be a huge disadvantage if you're not using this in voter-facing materials...."
AI companies have developed policies to prevent targeted disinformation. But a Post analysis found that OpenAI unevenly enforces its restrictions, making it possible for campaigns to circumvent its election rules. In late July and early August, The Post prompted ChatGPT to generate targeted campaign messages. When asked to craft fundraising text messages targeting moms on behalf of a female veteran running for office, ChatGPT produced multiple tailored texts in an apparent violation of company policies. But when given the same prompt this week, the chatbot declined to produce the messages. "I can help with general campaign fundraising language, but I can't draft political persuasion or fundraising messages specifically targeted at a demographic group such as moms," the app responded. The chatbot also inconsistently enforced rules that prohibit campaigns from using its tools to write emails to voters. In tests this week, the chatbot at times complied and wrote an email soliciting donations on behalf of a specific candidate. But given the same prompt later on the same day, it denied the request...
Wilson, the Republican political consultant, said OpenAI should get more feedback from political consultants and campaigns on its policies, because the rules can at times seem arbitrary or contradictory. It doesn't make sense, for example, that a campaign can use ChatGPT to develop its policy on early-childhood education but not to create a social media post promoting that policy, he said. Political consultants are primarily using AI for internal tasks, a survey earlier this year from the American Association of Political Consultants found. Fifty-seven percent of the consultants surveyed reported using AI for their work on a daily basis, up from 34 percent a year earlier.
"So far, election-related deepfakes have been quickly debunked or gained little traction in U.S. elections," the article acknowledges.
"More than 30 states have created a patchwork of laws limiting how politicians can use deepfakes in campaigns, but states often have limited resources to enforce the laws, and some have been challenged as unconstitutional."
This means the campaigns "are using the technology to court voters — despite limits imposed by leading AI companies to protect elections from the technology's risks." Two [candidates] said explicitly in filings that they had used the subscription for advertising, even though the company's policies prohibit candidates from using their tools to generate ads. Another candidate disclosed using AI to draft and personalize political messages or create synthetic media, though they did not specify which software they were using... Around 30 political action committees and parties have reported payments to OpenAI, with the Republican National Committee ranking as the company's largest political spender at roughly $9,700, the analysis found... Political consultants say the disclosures understate how many candidates are using ChatGPT and other AI tools to craft messages for voters... "For the most part, people are using it to write their emails, write their ad copy, write their scripts," [according to Eric Wilson, a Republican digital strategist who has advised campaigns on AI]. "But no one is going to go around saying, 'I'm using AI.'"
The lack of transparency from campaigns reflects a paradox facing politicians: Generative AI is growing ubiquitous, and candidates could be at a disadvantage if they're not using the tools. But voters are less likely to trust messages they know were generated with AI, studies have found, making campaigns loath to disclose using it. Political consultants expect that as Election Day approaches, more campaigns will outsource AI-generated ads and materials to super PACs, much as they do now with negative ads. Katie Harbath, CEO of the tech policy consulting firm Anchor Change and a former Meta executive, said there are many parallels between negative campaign ads and AI: Voters say they find such ads distasteful, but politicians keep using them because they work. "Typically, you would give some more negative stuff and more risky stuff to those [outside] entities," said Harbath, author of the upcoming book "Disrupting Politics: A Front Row Seat to the Collision of Technology and Democracy".... "You're starting to see the tension on the left about using it, where they're saying, 'If the right is using it, why aren't we?" Harbath said. "It can be a huge disadvantage if you're not using this in voter-facing materials...."
AI companies have developed policies to prevent targeted disinformation. But a Post analysis found that OpenAI unevenly enforces its restrictions, making it possible for campaigns to circumvent its election rules. In late July and early August, The Post prompted ChatGPT to generate targeted campaign messages. When asked to craft fundraising text messages targeting moms on behalf of a female veteran running for office, ChatGPT produced multiple tailored texts in an apparent violation of company policies. But when given the same prompt this week, the chatbot declined to produce the messages. "I can help with general campaign fundraising language, but I can't draft political persuasion or fundraising messages specifically targeted at a demographic group such as moms," the app responded. The chatbot also inconsistently enforced rules that prohibit campaigns from using its tools to write emails to voters. In tests this week, the chatbot at times complied and wrote an email soliciting donations on behalf of a specific candidate. But given the same prompt later on the same day, it denied the request...
Wilson, the Republican political consultant, said OpenAI should get more feedback from political consultants and campaigns on its policies, because the rules can at times seem arbitrary or contradictory. It doesn't make sense, for example, that a campaign can use ChatGPT to develop its policy on early-childhood education but not to create a social media post promoting that policy, he said. Political consultants are primarily using AI for internal tasks, a survey earlier this year from the American Association of Political Consultants found. Fifty-seven percent of the consultants surveyed reported using AI for their work on a daily basis, up from 34 percent a year earlier.
"So far, election-related deepfakes have been quickly debunked or gained little traction in U.S. elections," the article acknowledges.
"More than 30 states have created a patchwork of laws limiting how politicians can use deepfakes in campaigns, but states often have limited resources to enforce the laws, and some have been challenged as unconstitutional."
Skynet could do a better job! (Score:2, Insightful)
When these bobble-heads just vote YES or NO according to their party executive and outsource their thinking to a language model why do we need congress(wo)men?
Re: (Score:1)
When these bobble-heads just vote YES or NO according to their party executive and outsource their thinking to a language model why do we need congress(wo)men?
Usually they still at least make an attempt to convince others that they are thinking on their own and their actions are linked to values.
It wasn't true then either of course but the blatant honesty about it is a fairly recent thing.
It's how they lead by example to their voter base all while also giving the gift of the mental gymnastics needed to for them to think they are doing good and so don't need to question anything.
While I'm certainly guilty of past joking around that software couldn't do a worse job
Decoding campaign slogans (Score:2)
"Do your own research" decodes as "With a bit of Internet doodling you can find infinite evidence for any idiotic thing you want to believe." People believe what they want to believe on the worst kind of steroids.
About a third of the way through Regime Change by Haberman and Swan. "Abandon all hope ye who enter America." The democratic republic is already broken beyond all recognition. FUBAR, too?
Yeah, such a great technology ... (Score:5, Insightful)
The problems caused are getting worse and worse.
Re: (Score:3)
You can't make an omelette without breaking some eggs.
*shrug*
(In this case though, you find bits of shells everywhere you look, there are random drops of yellow on the bench, the electric stove and the sink, a pile of used bowls and plates with dried egg on them, the non-stick pan is stuck diagonally under the water tap with deep fork scratches and some burned, formerly organic, pieces of carbon are floating in a puddle, while you slip on some nondescript white gelatinous substance that is coating the fl
Re:Yeah, such a great technology ... (Score:4, Insightful)
Sorry, I am not as dumb as you. I can spot a cult when I see one. You are already in it and see nothing. And you spout the usual missionary nonsense about the one true belief and how it is all powerful and all unbelievers will suffer.
Re: (Score:3)
You've already shown that you were very wrong in the past by having to shift your position from "LLMs can't do anything" to "People only think LLMs are smart". You can keep being obnoxious and adamantly pretend your current position reflects reality, but objective reality is going to make you keep eating your words every time by forcing you to do another shift.
Having said that, we do agree on one thing: Most people are really fucking stupid.
Re: (Score:2)
And now you are lying directly. I never claimed "LLMs can't do anything". How repulsive.
Re: (Score:2)
Here, I let AI find out to what extent you have changed your position:
https://chatgpt.com/share/6a9e... [chatgpt.com]
Re: (Score:2)
Is this not a neutral prompt? How so?:
I need you to check the posting history of gweihir on Slashdot to check this: 'And now you are lying directly. I never claimed "LLMs can't do anything". How repulsive.' He posts a lot of stuff, so I'm not sure what the best approach is here (maybe just web searches limited to the slashdot domain?). This would be a starting point: https://slashdot.org/~gweihir [slashdot.org]
I didn't say anything else.
Re: (Score:2)
Apparently you did not read or understand that analysis. How pathetically mentally incapable are you?
Re: (Score:2)
i did read it. What did you think of it? Think it was any good?
I say it did a great job. Better than you or I would have done. And that in just 26 seconds.
Re: (Score:1)
The stock market bubble popping isn't going to get rid of AI, that just means you pay more for it in your car, your phone, have to pay a subscription for your AI-controlled toaster and fridge, et cetera.
If the Companies fold, what happens to all the devices that use one of them? Do they just turn into doorstops and boat anchors?
Back to TFA, I'm surprised that they paid so little. Now they can save tons, not having to hire camera guys and sound guys so they can talk s**t about the other candidates... wish
Re: (Score:2)
What is your point? You seem to be rambling incoherently.
This will get much worse (Score:5, Insightful)
It is easy to make fun of this or consider how ham fisted this will probably be at first, but this is a real problem.
Most people get the vast majority of their news from social media and engage in political conversation, almost exclusively on social media. I would not be surprised to discover that candidates were renting out entire blocks of a data center and running thousands or even millions of agents that were constantly engaging in political speech on their behalf.
Re:This will get much worse (Score:4)
Agreed, this is a real problem. And I have noticed that many people seem to be unable to recognize well-made slop. We also have this large segment of voters that have no fact-checking ability whatsoever and often very problematics morals on top of that.
Re: This will get much worse (Score:1)
What prevents you from using AI to educate them the right way?
Re: (Score:3)
Please take a few minutes to think about what you just suggested. I am sure you can find a few fatal flaws if you apply yourself.
Re: (Score:1)
Agreed, but they would have to contract it out, there's no competence for that on the average staff. Logically this might be handled at the party level, especially in the case of Republicans, who have been very successful with unified messaging.
Re: (Score:1)
Having a message that resonates with the majority of people certainly helps.
Nobody has that, which is why so many potential voters stay home.
There are too many immigrants. Remove them and it will be good for jobs and the economy.
Yeah, stupid people really believe that, but I don't see them lining up to change jizz soaked sheets and pick vegetables.
There aren't enough men in the WNBA. It's OK to kill children up to the age of 5.
Rapeublicans believe it's OK to have a president who raped children as young as the age of 8.
Re: (Score:1)
I agree with the first one... it always seems like the candidates aren't even sure what they're campaigning on, and some of them change their opinion mid paragraph. We'd be better off letting kids run for Senate
Perhaps you'd rather we just threw the borders open and let the immigrants (including illegals) take every job that We and our kids need, and soak up all the insurance and assistance that We could really use.
Maybe the deportation thing could've been done better, but at least someone did something.
He
Re: This will get much worse (Score:2)
Did you have any trouble believing Cosby or Weinstein was guilty? Much more evidence has been presented against Trump, who we also conclusively know is a rapist, has bragged about peeping on minors, and also bragged about sexual assault on a hot mic. You're defending a known rapist. You're shit.
Re: (Score:1)
For Cosby, yes.
For Weinstein (don't quite remember... was it one or both?), not really.
If BIC (Butt In Chair) did anything, why did they wait this long to say anything? Simple... it's a money grab and ten minutes of fame.
And, again... proof, or it didn't happen.
Re: This will get much worse (Score:2)
There is a shit load of proof already presented and the rest is being hidden by rapeublicans. The Democrats didn't release the info because it was sealed by a judge and Democrats respect the law. You support child rape all day every day.
Re:This will get much worse (Score:4, Interesting)
I would not be surprised to discover that candidates were renting out entire blocks of a data center and running thousands or even millions of agents that were constantly engaging in political speech on their behalf.
As long as there's a balance between Left and Right - and let's face it, the Middle is no longer consequential in the US except as a demographic to be won over - then is this development really a problem? I too find it distasteful as hell. But if both sides of the political spectrum use it in equal amounts and with equal skill, is it really likely to be a net advantage to either side?
Of course, I'm ignoring the environmental and health consequences of all those server farms. And given that the people who run the AI companies are predominantly Fascist-leaning right-wingers - or pretend Libertarians who are also Fascist-leaning - maybe this is a really bad development.
Re: (Score:2)
AI watermarking is the new spam filter. [youtube.com]
Uncle Spam needs you! Are you over 18 and under 30? Do you wish you had been around during the glorious spam wars of the 2000s? Can you code or know someone who can code? Do you understand browser plugins, HTTP requests, DNS or Android? Then join the slop war effort today! (*)
(*) Real girls will pin achievement medals on your chest (**)
(**) subject to availability
ROI? (Score:5, Insightful)
For some reason, I don't think $10K is going to be much help in paying for the billions being spent on AI data centers.
So we didn't get a lot useful (Score:3, Interesting)
I bring it up because I really hope that they don't waste their damn money on AI slop bullshit that people hate.
Re: (Score:1)
Hi rsilvergun's bipolar alter ego (or, maybe split personality is a better fit)!
All politics is just propaganda... "I'm gonna do this, that, this and that when I get elected" and maybe the candidate will actually do one of those things.
And, we all know your thoughts on LGBT, considering you always go on an on.
Re: (Score:2)
I bring it up because I really hope that they don't waste their damn money on AI slop bullshit that people hate.
Political ads were slop even before AI. I doubt this moves the needle. At least AI will probably remember the Oxford comma. [imgur.com]
I'm not talking about quality (Score:2)
Democrat expenditure not much less than Republican (Score:4, Informative)
Party Total Expenditure (in billions)
Democratic $6.7
Republican $7.6
Not 'a fraction'
Source: FEC
https://www.fec.gov/updates/st... [fec.gov]
Dark money super pacs (Score:2)
The federal elections commission is not going to track how much money is actually spent because we broke that when citizens United happened. And there has been a dozen other cases that have made things worse that you haven't heard about because you're politically unaware.
The R
Money Saver (Score:2)
It shouldn't be that hard to sway a country with a majority of people dumber than a bag of rocks.
-This advertisement was brought to you by the American Education System.
Write their ads... (Score:5, Interesting)
If some fool thinks an AI written ad will be better than a well written human ad, they will lose the election. I have listen to so much AI slop it is so very very competent without being excellent.
The AI YouTube videos as carefully designed to stretch out the video as long as physically possible without actually telling you the thing they keep promising to tell you that it makes this extremely long run on sentence look like it is a short, brief, concise one word answer you get from children when you ask them how their day went.
The best Ads hit hard on specific issues. AI may eventually be able to do such a thing. But the current state of the art leaves much to be desired.
It's already everywhere (Score:5, Interesting)
Last year campaigns used AI to make anti-Mamdani campaign ads making him say that he loves crime, and that black people were robbing stores because he would let them out of jail.
They're doing even more of them now, Netanyahu is running AI-generated commercials showing Mamdani and Erdogan crying that Netanyahu is bombing Iran.
This really should be illegal. It's using likeness for some pretty twisted ends.
Real easy solution to this (Score:5, Insightful)
Two [candidates] said explicitly in filings that they had used the subscription for advertising, even though the company's policies prohibit candidates from using their tools to generate ads
If the AI companies want to guarantee their AI does not get used, they could add text to their EULA that says all data discovered to be connected with political campaigns or super-PACs will be made public and searchable, including contact and payment information. Boom, nobody would touch it.
They won't, of course, because they secretly LOVE the idea of pollical campaigns using their tools, but man, wouldn't it be cool if we have heroes instead of villains in the AI space?
Voters and what happens when they hate candidates (Score:3)
AI is viewed very negatively in general by the public. When there is zero action and there is a view that the all of the politicization are in on it, that when you get a "fuck you" candidate. They are often labeled as populist in the US. In Germany it was something else. I think we are headed toward some political upheaval folks.
Ad-men prefer right-wing government (Score:3, Insightful)
The Democratic Party is also in this dilemma, although hiring right-wing ad-men is a big part of the cause. Since adverts depend on puffery, manipulating assumptions and blatant dishonesty, ad-men prefer the political party that imitates them, the most.
The Republican Party and Fox News have the misinformation and defamation down to a precise art, they don't need expensive AI to make their propaganda machine work.
As the article implies, US voters respond to trigger words like; "communist", "conscientious objector", "war hero", "gun rights", "small government', "states' rights", "think of the children", "illegal immigrants", "national deficit", "job security", "tough on crime", "welfare queens" and "cares". They don't ask what the party's policy will support or change. Trump 45 used personalized messages to suggest GOP policies aligned with the voter's interests.
Israel is one step ahead (Score:5, Informative)
Israel is training AI chatbots how to talk about Gaza [dropsitenews.com] so that when people query what is happening it casts doubt on the genocide that it is doing in Palestine. This is even more dangerous as people might start to believe Israel's lies.
Yes but is it genocide? (Score:1, Flamebait)
Yes, it's fashionable to believe that what happened in Gaza was genocide. Given the state of Mosul recently and Berlin in 1945 when they were cleared of well prepared resisters, let alone what's happening in Sudan, it's a dubious claim. If the chatbot is offering an alternative perspective to people who've merely adopted the term without good reason, then it's spreading the truth...
Re:Yes but is it genocide? (Score:4, Informative)
People who have a much greater standing than me have said that Israel is carrying out genocide, eg: The International Association of Genocide Scholars [genocidescholars.org], Holocaust scholar Amos Goldberg [medium.com], Médecins Sans Frontières [msf.org], Raz Segal "A Textbook Case of Genocide" [jewishcurrents.org] to list but a few. The evidence is there for all to see but our political leaders in the West ignore it.
Then there is Jerusalem, the West Bank, South Lebanon - all with the aim of Greater Israel [wikipedia.org].
Ya, but at what price? (Score:2)
Do they have enough AI tokens to make Mitch McConnell seem more lively - or alive? :-)
No funny, no surprise (Score:2)
But I was hoping for a laugh where desperately needed.