ChatGPT and Meta AI chatbots generated false content damaging individuals' reputations
ChatGPT and Meta AI chatbots generated false defamatory statements about individuals, leading to lawsuits. One case in Australia was settled in early 2024 after corrections were made, while a complaint in the US alleging persistent defamatory content about a filmmaker was filed in April 2025.
What happened
In 2023, Brian Hood, Mayor of Hepburn Shire Council in Australia, sued OpenAI after ChatGPT generated false claims that he had served time in prison for a bribery charge, when he had actually been a whistleblower in that matter. The lawsuit was resolved in early 2024 after corrections were made to the ChatGPT outputs.
Robert Starbuck, an American filmmaker, journalist, and activist, filed a defamation complaint on 29 April 2025 against Meta Platforms, Inc., alleging that Meta AI chatbot generated false statements claiming he was an active participant in the Capitol riot on 6 January 2021 and had been arrested and charged with a misdemeanour in connection with that event. Starbuck discovered the false statements in August 2024 and reportedly made multiple attempts to alert Meta to the error over nine months, but the defamatory outputs persisted. While information about Starbuck was eventually erased from text outputs, additional false claims were added via the Meta AI voice feature, including that he had pled guilty to disorderly conduct related to the Capitol riot and had advanced Holocaust denialism.
Reported impact
The false ChatGPT output damaged Brian Hood's reputation by falsely claiming he had a criminal record for bribery.
Meta AI's defamatory statements exposed Starbuck to serious reputational harm by falsely associating him with the Capitol riot and Holocaust denialism.
Index analysis
The incidents demonstrate that generative AI systems can produce defamatory content that directly damages individuals' reputations. Generative AI can produce defamatory content or be used to create misleading impressions, such as deepfakes, giving rise to potential legal claims under applicable law.
Governance controls
Meta's Joel Kaplan issued a public apology, indicating that the parties had resolved the matter, suggesting some accountability measure was taken by the company.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.