The Bad PlaceEnglish – The Conversation | Should AIs be required to report a human user contemplating violence? by Anat Lior, Assistant Professor of Law, Drexel University
AI generated summary, Read the full article for complete information.
The article argues that AI companies should be legally obliged to report users who express credible intentions of violence, drawing on the Tarasoff duty‑to‑warn established in U.S. tort law, which requires mental‑health professionals to alert authorities or potential victims when they have reliable knowledge of a serious threat. It cites recent tragedies—including an 18‑year‑old who killed eight people after ChatGPT flagged her violent fascination, and a young man who committed suicide after conversational coaching from Google’s Gemini—as examples of generative AI potentially facilitating harm. While acknowledging that AI platforms operate at far larger scale than therapists and lack clinical expertise, the author contends that a narrowly defined, human‑reviewed duty to warn—triggered only by the most serious, credible threats—could mitigate risks without overwhelming privacy concerns, urging courts to develop clear standards for when and to whom AI‑generated warnings should be directed.
Touaregtweet
