Do AI companies have a duty to warn?

Plus: Pope Leo's encyclical on artificial intelligence ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌
The Conversation

Top headlines

Lead story

In most states, if a patient in a therapy session threatens to harm someone, the therapist has a legal duty to notify authorities and ensure that the intended target is warned. Meanwhile, many people turn to AI chatbots for mental health support – or to get advice on how to commit acts of violence. This raises an important question: Should AI companies also have a duty to warn?

Several high-profile cases of people killing others or themselves after interacting with chatbots show that this is not an abstract question. AI companies sometimes find themselves with the same information that would require a human therapist to warn authorities.

Drexel University law professor Anat Lior explains that the therapist’s duty to warn could be a model for AI companies, but also lays out the challenges of taking the practice from human–human interactions to interactions between millions of people and a set of computer algorithms.

[ Sign up for our weekly Global Economy & Business newsletter, with expert perspectives from around the world. ]

Eric Smalley

Science + Technology Editor

Students hold a vigil near the scene of a shooting at Florida State University. The gunman allegedly consulted ChatGPT about how to carry out the attack. Miguel J. Rodriguez Carrillo/Getty Images

Should AIs be required to report a human user contemplating violence?

Anat Lior, Drexel University

Human therapists have a legal duty to warn authorities and potential targets when patients say they plan to harm someone. The same can – in theory – be required of AI chatbots .

Pope Leo XIV attends the presentation of his first encyclical, Magnifica Humanitas, at the Vatican on May 25, 2026. AP Photo/Alessandra Tarantino

Pope Leo XIV compares AI to the Industrial Revolution – as new alternatives to big AI firms take shape

Nathan Schneider, University of Colorado Boulder

Leo XIV released his first encyclical on the 135th anniversary of Rerum Novarum, the 1891 papal document on the upheavals of the Industrial Revolution.

Health + Medicine

Economy + Business

Arts + Culture

Politics + Society

Environment + Energy

Trending on site

Today's graphic 📈