Senate Republicans Deploy AI-Generated Deepfake of Chuck Schumer in Controversial Attack Ad

Senate Republicans deepfaked Chuck Schumer, and X hasn't taken it down | TechCrunch

Senate Republicans shared a deepfake video of Chuck Schumer, the Senate minority leader, designed to make it seem like Democrats are celebrating the ongoing government shutdown, which has lasted 16 days. In the deepfake, an AI-generated Schumer repeats the phrase "every day gets better for us," an actual quote taken out of context from a Punchbowl News article. In the original story, Schumer discussed the Democrats' a healthcare-focused shutdown strategy, and said they were not going to back away from Republicans' playbook of threats and "bambooz[ling]." The shutdown is happening because Democrats and Republicans cannot agree to pass a bill funding government through October and beyond. Democrats are trying to hold onto tax credits that would make health insurance cheaper for millions of Americans, secure a reversal to Trump's Medicaid cuts, and block cuts to government health agencies. The video was posted Friday on the Senate Republicans' X account. According to X's policies, the platform prohibits "deceptively shar[ing] synthetic or manipulated media that are likely to cause harm." Harmful content includes media that could "mislead people" or "cause significant confusion on public issues." Enforcement actions include removing content, labeling warnings, or reducing visibility. X has not, as of the time of this writing, removed the deepfake or added a warning label -- though the video does include a watermark denoting its AI origins. The Schumer video is not the first time X has allowed deepfakes of politicians to remain on the platform. In late 2024, X owner Elon Musk shared a manipulated video of former Vice President Kamala Harris in the lead up to the election, sparking debate about misleading voters. TechCrunch has reached out to X for comment. Up to 28 states have enacted laws prohibiting deepfakes of political figures, specifically around campaigns and elections, though most don't outright ban them if they have clear disclosures. California, Minnesota, and Texas have banned deepfakes intended to influence elections, deceive voters, or harm candidates. The latest post comes weeks after President Donald Trump posted deepfakes on Truth Social depicting Schumer and Hakeem Jeffries, the House minority leader, making false statements about immigration and voter fraud. Responding to criticism of the lack of honesty and ethics, Joanna Rodriguez, the National Republican Senatorial Committee communications director, said: "AI is here and not going anywhere. Adapt & win or pearl clutch & lose."

A GOP Account Posted a Chuck Schumer Deepfake With a Real Schumer Quote

Is an AI-generated video of a politician technically a deepfake if the words they’re saying are their own? By any reasonable definition, yes, and the National Republican Senatorial Committee published just such a deepfake of Democratic Minority Leader Chuck Schumer on its X account on Friday: The ad, which is labeled as AI-generated, depicts Schumer saying six words he did actually say: “Every day gets better for us,†but the quote originally comes from a text interview with Schumer published in Punchbowl News. After saying these words in the video, Schumer’s digital mouth peels into a nightmare grin like Jim Carrey in The Grinch. “These are Chuck Schumer’s own words,†NRSC communications director Joanna Rodriguez told The Associated Press, adding that “video is a way for voters to see and hear it themselves.†Reading the quote aloud in a scary voice and putting it onscreen in red text is a more traditional way of doing this in political ads. The video is vaguely cinematic, with heavy bokeh. Savvy internet content consumers probably would not mistake this for an off-the-cuff-remark taken from a TV news clip, or a smartphone video taken by a congressional reporter. It looks more like a stock photo of Schumer hastily animated with a quick-and-dirty online tool like Google’s Veo 3. That’s not to say there’s nothing to worry about. According to a famous genre of viral video, Facebook users in their 60s and older fall for much less plausible AI trickery all the time. The X rules state “You may not deceptively share synthetic or manipulated media that are likely to cause harm.†The “likely to cause harm†phrasing leaves room for interpretation, particularly since the video has only received 364 reposts and 414 likes on X at the time this was published. In the original interview, Schumer is essentially saying his party’s messaging around the government shutdownâ€"an attempt to focus the public’s attention on healthcareâ€"is going well. He even reiterates this in slightly longer form: “But every day we’re getting better and better as the message sinks in more and more deeply.†We reached out to Senator Schumer for comment, and will update if we hear back.

A GOP attack ad deepfakes Chuck Schumer with AI

The video posted by the National Republican Senatorial Committee shows an AI-generated version of Democratic Senate Minority Leader Chuck Schumer repeatedly saying "Every day gets better for us" and grinning. @NRSC/Annotation by NPR hide caption A new attack ad from Senate Republicans uses Sen. Chuck Schumer's real words about the government shutdown -- but in an AI deepfake of the Democratic Senate Minority Leader. The 30-second video posted on X and YouTube by the National Republican Senatorial Committee on Friday raised alarms among many observers who warned it crossed a new boundary in politics and could unleash a flood of AI-generated deepfake attack ads. Captioned on X "Week 3 of the Schumer Shutdown: 'Every day gets better for us'," the video shows an AI-generated version of Schumer repeatedly saying those words and grinning. The video concludes with Schumer smiling as a narrator says, "The Schumer shutdown is making things worse across America and Democrats love it." "Every day gets better for us" was part of a quote Schumer gave to the politics site Punchbowl News in an Oct. 9 article about the shutdown. He continued: "It's because we've thought about this long in advance and we knew that health care would be the focal point on Sept. 30 and we prepared for it ... Their whole theory was -- threaten us, bamboozle us, and we would submit in a day or two." However, the original story only included the text of Schumer's statement, not video or audio. An NRSC press release announcing the ad describes it as "visualizing" Schumer's comment and says it was "created in-part using artificial intelligence software and includes an AI disclaimer." The fake footage of Schumer includes a small transparent watermark of the NRSC logo and the words "AI GENERATED" in the bottom right-hand corner. YouTube put its own "Altered or synthetic content" label on the video saying "Sound or visuals were significantly edited or digitally generated." Schumer's office did not respond to a request for comment. Even though the quote was something Schumer did in fact say, the use of AI to invent fake video footage crosses a line, said Hany Farid, a professor at the University of California at Berkeley who studies manipulated media. "If they wanted to use the quote then they could have shown an image of Schumer and overlaid the quote, as is often done. It cannot reasonably be argued that creating a deepfake of Senator Schumer is the same thing," he said. In addition, Farid said that the disclaimer on the lower right corner of the video "is not a particularly clear or obvious disclosure, particularly when most users are doom scrolling through their social media posts." The NRSC video was also criticized by journalists and other commenters on social media, including reporters at right-leaning publications. "The Schumer bit is AI generated video -- which is disclosed at bottom right corner -- but we are at the doorstep of a terrifying new world," Jon Levine, a political reporter at the Washington Free Beacon, wrote on X. "NRSC has increasingly leaned into AI for its ads, but this is the first time I've seen them go so far as to use it for a fake video of an opponent. This is a slippery slope, even w/the tiny AI disclosure in the corner," Ramsey Touchberry, a congressional reporter at the Washington Examiner, posted on X. Amid the criticism, NRSC Communications Director Joanna Rodriguez wrote on her own X account: "AI is here and not going anywhere. Adapt & win or pearl clutch & lose." It's not the first time Republicans have used AI to impersonate Schumer and other Democrats. On the eve of the shutdown, President Trump posted on Truth Social an apparently AI-generated, profanity-laced video fabricating Schumer saying "nobody likes Democrats anymore." House Minority Leader Hakeem Jeffries is also shown in the video wearing a cartoonish sombrero and mustache. But unlike the NRSC's video of Schumer, earlier examples were easy to spot as inauthentic. Farid warned that the growing prevalence of AI fakes in politics could have a corrosive effect. "While I don't think our leaders should be posting deceptive deepfakes, they also run the risk that when they post real content, there will be good reason for the public to question its authenticity," he said.

Republicans use deepfake video of Chuck Schumer in new attack ad

It is not the party's first foray into the dystopian territory, with Trump having posted AI videos of his own The National Republican Senatorial Committee crossed into dystopian new territory for political campaigning on Friday after releasing an attack ad that features an artificially generated video of Senate minority leader Chuck Schumer. The deepfake video, posted Friday to the Senate Republicans' social media account, shows an AI-generated Schumer robotically repeating the phrase "every day gets better for us" in reference to the ongoing government shutdown. A small disclaimer tucked in the corner acknowledges its artificial origins. "Week 3 of the Schumer Shutdown: 'Every day gets better for us,'" the post read, accompanied by the deepfake video looping the Democratic leader's manufactured likeness. The video has bewildered those who watched it online, given that the quote itself is real and on the record for Punchbowl News. In the original interview, Schumer explained that Democrats had prepared their healthcare-focused shutdown strategy well in advance, adding: "Their whole theory was - threaten us, bamboozle us and we would submit in a day or two." Yet rather than simply quote Schumer's words, the NRSC, the campaign arm responsible for electing Republican senators and chaired by Tim Scott, a South Carolina senator, chose to manufacture synthetic video of him speaking. "Schumer thinks playing with Americans' livelihoods is just a game," the narrator said over the deepfake imagery. The advertisement goes on to feature other discussion on the shutdown, accusing Democrats of "loving" the political standoff. In a response on X, Joanna Rodriguez, the NRSC communications director, said: "AI is here and not going anywhere. Adapt & win or pearl clutch & lose." The video comes weeks after Donald Trump posted his own AI-generated deepfake to Truth Social, depicting Schumer and Hakeem Jeffries, the House minority leader, making crude and made-up statements about immigration and voter fraud that neither lawmaker uttered.

Senate GOP committee shares AI-made video of Schumer reveling in government shutdown

Why it matters: There's a lot more of this coming fast in the AI era. * Punchbowl News reported the quote from Schumer on Oct. 9 based on an interview with Schumer on Capitol Hill. * Schumer was said to be describing the why he thought the political momentum was in Democrats' favor on the shutdown. Our thought bubble: No laws ban what the NRSC did here, though there is bipartisan support for the type of watermarking seen in the bottom right corner of the video. * If this were an ad, the tiny amount of FEC regulation around digital election advertising would not have banned this sort of thing. * It marks a new normal in how organizations are willing to use AI to spread their messages. The big picture: Since the release of OpenAI's Sora app last month, realistic deepfakes are flooding social media. One deepfake debunker is making it his mission to explain how to spot AI video as a key part of media literacy for everyone. * "If you're tricked by an AI possum eating Halloween candy, that doesn't mean you're stupid," Jeremy Carrasco told Axios in an interview last week. * "That's a learning opportunity for when the politician is being deepfaked by AI." The bottom line: It's not just that a lot more deepfakes are coming, it's that they're looking a lot more real.