California state Sen. Scott Wiener reversed course and shut down a parody AI chatbot targeting his congressional opponent, one day after Nancy Pelosi and a group of female Democrats condemned it as sexist.
Wiener's campaign had built a website called ConnieChan.ai that featured a chatbot impersonating San Francisco Supervisor Connie Chan, his rival for Pelosi's old congressional seat. The bot greeted visitors with a taunt: "I'm Connie, the world's first AI trained to say 'no' to anything you ask." A second line added: "Bring me an idea and I'll explain why it can't, shouldn't, and won't happen." Wiener's campaign also put up billboards around San Francisco promoting the site.
The stunt lasted until Pelosi weighed in. The former House speaker posted on X condemning the chatbot as "sexist." Rep. Zoe Lofgren and eight other female Democrats from California, New York, Washington, and Hawaii backed her up. By Friday, Wiener folded.
The reversal was swift, and telling. On Thursday, Wiener's initial response was combative. He accused Chan's campaign of trying to deflect from her "awful record of obstruction and inaction" and claimed her team had "bombarded the chatbot with repeated malicious questions." He called it "ironic" that Pelosi would raise concerns about AI, a reference to Pelosi's lobbying against Wiener's first major piece of AI legislation, which Gov. Gavin Newsom vetoed in 2024.
That defiance evaporated within twenty-four hours. Politico reported that Wiener issued a written statement Friday announcing the chatbot's removal and an apology for the offense it caused.
"My campaign created ConnieChan.ai as a parody to draw attention to my opponent's record of delay and obstruction around housing, public transit, and other critical needs."
That was Wiener's framing. But his next lines read like damage control, not conviction:
"Many others were offended, and I apologize for causing that offense. Innovation means trying new things, and it also means being honest when something doesn't land the way you intended."
He confirmed the chatbot was coming down immediately and said the billboards would follow "soon." No specific timeline for the billboard removal was given.
Connie Chan's team was not satisfied. Her campaign had argued the chatbot's responses carried racist and sexist connotations, including references to Chan's accent and citizenship status. Chan campaign spokesperson Ian Krager fired back Friday:
"[Wiener can] rightly take down his bigotry bot, but he can't erase the concerns it raised among San Franciscans about his character and judgment."
Krager also accused Wiener's campaign of "mocking an immigrant woman of color." The specific chatbot responses that prompted those accusations were not reproduced in available reporting, so it remains unclear exactly what the bot said about Chan's background when prompted.
Pelosi endorsed Chan as her successor to represent San Francisco in Congress. The race has become one of the most closely watched intra-party contests in the 2026 cycle, pitting a progressive state legislator who has positioned himself as an AI policy leader against a Pelosi-backed local official.
The episode carries a particular irony for Wiener, though his campaign would rather not dwell on it. He has made artificial intelligence regulation a signature issue in Sacramento. Newsom vetoed his first major AI bill in 2024, but signed a second Wiener-authored AI measure in September 2025.
A lawmaker who spent years crafting rules for responsible AI deployment chose to use the same technology to mock a political opponent, then pulled the plug the moment powerful allies objected. That sequence says more about Wiener's instincts than any policy paper.
Wiener has already faced pressure from multiple directions in this race. He struggled to satisfy anti-Israel activists on the progressive flank earlier in the campaign, and the chatbot fiasco now hands his opponents fresh ammunition on the question of temperament.
Despite those headwinds, Wiener has led in early congressional primary polling even as Pelosi's preferred candidates lagged. The chatbot controversy tests whether that lead can survive a direct clash with the most powerful figure in San Francisco Democratic politics.
What stands out is the speed of the political response. Pelosi posted her condemnation on Thursday. Within the same day, ten female members of Congress, Lofgren plus eight others from four states, had lined up behind her. By Friday morning, Wiener was apologizing. The whole arc, from defiance to surrender, took roughly a day.
That timeline matters. Wiener did not remove the chatbot because he reconsidered its merits overnight. He removed it because Pelosi made it politically untenable to keep it up. His own Thursday statement showed he was ready to fight. His Friday statement showed he was not willing to fight Pelosi.
This is the kind of intra-party discipline Pelosi has exercised for decades, and it worked exactly as designed. A sitting state senator with a legislative record on AI policy, someone who should know better than most how these tools land in public, got yanked back into line in under twenty-four hours.
Several questions remain unanswered. What exactly did the chatbot say when users asked about Chan's accent or citizenship? Who approved the bot's deployment within Wiener's campaign? And how does a legislator who has spent years writing rules for AI use justify deploying a parody bot that his own allies called bigoted?
The contest for Pelosi's San Francisco seat is part of a broader pattern of intra-Democratic fights playing out across the 2026 primary map, where establishment figures and progressive insurgents are clashing over direction, tone, and tactics.
Wiener's chatbot episode is a small story on its own, a campaign gimmick that backfired. But it captures something larger about the progressive political class: eager to regulate everyone else's use of powerful tools, and remarkably reckless with those same tools when their own ambitions are on the line.
