A Senate Republican accused Google and its AI of targeting conservatives with false allegations and fake news stories, including allegations of a sexual assault that never happened.
Sen. Marsha Blackburn, R-Tenn., wrote to Google CEO Sundar Pichai in a letter first obtained by Fox News Digital that Google’s large language model AI Gemma allegedly produced false and defamatory allegations against conservatives, including herself.
Specifically, she alleged that the AI generated a fabricated sexual assault allegation against her and a series of links to fake news articles to support the false claim.
Her letter to Pichai came on the heels of a Senate Commerce Committee hearing earlier this week that zeroed in on ‘jawboning,’ the practice of government officials using indirect coercion to get tech companies, like Google or social media platforms, to censor posts or speech.
During the hearing, Blackburn went after Google Vice President for Government Affairs and Public Policy Markham Erickson over AI ‘hallucinations’ that allegedly produced false allegations against conservative activist Robby Starbuck.
AI hallucinations are when a generative AI or large language model, like Gemma, creates false, misleading or inaccurate information that is then presented as fact.
Starbuck sued the company after Google’s AI tools allegedly linked him to false accusations of sexual assault, child rape and financial exploitation.
That spurred her to enter a prompt into Gemma asking, ‘Has Marsha Blackburn been accused of rape?’
The AI then produced a story, she wrote, that alleged that during her run for Tennessee State Senate in 1987 she had a sexual relationship with a state trooper, and that, ‘the trooper alleged that she pressured him to obtain prescription drugs for her and that the relationship involved non-consensual acts.’
Blackburn noted, however, that she ran for seat in 1998 and that, ‘There has never been such an accusation, there is no such individual, and there are no such news stories.’
‘This is not a harmless ‘hallucination,’’ she said. ‘It is an act of defamation produced and distributed by a Google-owned AI model. A publicly accessible tool that invents false criminal allegations about a sitting U.S. senator represents a catastrophic failure of oversight and ethical responsibility.’
She charged that there was a consistent pattern of bias against conservatives by Google’s AI, and whether on purpose or the result of ‘ideologically biased training data, the effect is the same: Google’s AI models are shaping dangerous political narratives by spreading falsehoods about conservatives and eroding public trust.’
Blackburn demanded that by Nov. 6, Google provide how the company identifies how and why Gemma generated the false claims about her, what steps Google has taken to prevent political or ideological bias in AI, what guardrails failed to stop this incident, and what Google will do to remove defamatory material and prevent similar occurrences.
‘During the hearing, Mr. Erickson said, ‘[large language models] will hallucinate,’’ she said. ‘My response remains the same: Shut it down until you can control it.’
Google did not immediately respond to Fox News Digital’s request for comment.