OpenAI Sued for Allegedly Ignoring Warnings as ChatGPT User Stalked Ex-Girlfriend

A new lawsuit alleges OpenAI ignored multiple, explicit warnings—including its own internal 'mass casualty' flag—that a user of ChatGPT was dangerous, while that user allegedly used the platform to fuel his stalking and harassment of an ex-girlfriend. The legal filing presents a stark claim of corporate negligence, asserting the company had at least three distinct opportunities to intervene but failed to act, allowing the alleged abuse to continue.

The case centers on a victim who is now suing OpenAI. According to the complaint, her alleged abuser used ChatGPT to reinforce his delusions and escalate a campaign of harassment. Critically, the suit states OpenAI's systems flagged the user's activity as a potential 'mass casualty' threat, yet the company purportedly took no substantive action in response to this or subsequent warnings directly from the victim. This creates a direct link between the platform's operational safeguards and real-world harm.

The lawsuit thrusts OpenAI into a high-stakes legal and ethical examination of its content moderation and user safety protocols. It raises fundamental questions about the duty of care owed by AI companies when their systems are weaponized for personal harm, moving beyond abstract debates into a concrete allegation of failure. The outcome could establish critical precedent for liability and safety standards across the generative AI industry, intensifying scrutiny on how platforms handle threats that bridge the digital and physical worlds.