(Photo by Jakub Porzycki/NurPhoto via Getty Images)
A Pittsburgh man who was indicted for stalking and harassing multiple women allegedly received encouragement from OpenAI's ChatGPT.
The grand jury indictment of 31-year-old Brett Michael Dadig, announced today by the Justice Department, accuses him of targeting 11 victims across several US states this past year. Interestingly, prosecutors note that Dadig relied "on advice from an artificial intelligence chatbot," which the 21-page indictment reveals is ChatGPT.
"Dadig also discussed on his podcast how he used ChatGPT on an ongoing basis and that it was his 'therapist' and his 'best friend,' according to indictment, which was spotted by 404 Media. "According to Dadig, ChatGPT told him to continue to message women and to go to places where the ‘wife type’ congregates, like athletic communities."
That said, the indictment indicates ChatGPT’s support for Dadig involved praising his podcast, which focused on dating and "building a life worth chasing." The chatbot also advised him about meeting his future wife at a gym, rather than any outright endorsement of stalking. Even so, federal prosecutors claim in the indictment that "Dadig viewed ChatGPT's responses as encouragement to continue his harassing behavior."
The stalking allegedly includes Dadig showing up at the victims’ homes or businesses uninvited, “following them from their places of business, attempting to get them fired, taking and posting pictures of them online without their consent, and revealing private details (including their names and locations) online,” the Justice Department says.
Dadig also allegedly threatened his victims, "subjected at least one victim to unwanted sexual touching," prompting gyms to ban him.
OpenAI didn’t immediately respond to a request for comment. But it’s no secret that the company has been trying to prevent ChatGPT from engaging in sycophantic or flattering behavior, a feature that some users actually like.
It’s possible Dadig would have harassed these women even without consulting ChatGPT. The indictment notes that Dadig at one point posted on social media about being diagnosed with antisocial personality disorder and bipolar disorder.
Still, the case underscores concerns about AI chatbots contributing to or worsening unhealthy behavior, including delusions, among certain users. In October, OpenAI released research that found about "0.07% of users active in a given week," or around 560,000 users, exhibited possible "signs of mental health emergencies related to psychosis or mania.” Another 0.15% of the active weekly users showed signs of an emotional reliance on ChatGPT.
In the meantime, Dadig faces up to 70 years in prison and a fine of up to $3.5 million if he’s convicted of all charges, which includes violating two restraining orders.
Disclosure: Ziff Davis, PCMag's parent company, filed a lawsuit against OpenAI in April 2025, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.