Matthew Livelsberger used ChatGPT to plan Trump hotel explosion
Matthew Livelsberger used ChatGPT to plan Trump hotel explosion
Occurred: January 2025
Report incident 🔥 | Improve page 💁 | Access database 🔢
US Army soldier Matthew Livelsberger used ChatGPT and other generative AI tools to plan the explosion of his Tesla Cybertruck outside the Trump International Hotel in Las Vegas.
Investigations into Livelsberger's attack showed that he was looking for information on explosive targets, the speed at which certain rounds of ammunition would travel and whether fireworks were legal in Arizona.
Kevin McMahill, sheriff of the Las Vegas Metropolitan Police Department, called the use of generative AI a “game-changer” and said “This is the first incident that I’m aware of on U.S. soil where ChatGPT is utilized to help an individual build a particular device.”
Livelsberger committed suicide and initiated an explosion of several fireworks mortars that were in the bed of his rented Cybertruck. Seven people were injured and the hotel's exterior was damaged.
A general purpose tool that can be used for just about anything, ChatGPT is supposed to have safeguards that stop it producing responses that could result in harm.
In response to the incident, OpenAI said it was committed to seeing its tools used “responsibly” and that they are designed to refuse harmful instructions.
“In this case, ChatGPT responded with information already publicly available on the internet and provided warnings against harmful or illegal activities. We’re working with law enforcement to support their investigation,” a statement emailed to news organisations said.
The incident highlights how easily generative systems such as ChatGPT can be used to plan terrorist attacks and other events with public safety implications.
It also raises questions about why how generative AI could facilitate harmful intentions and emphasises the need for responsible use of such tools.
Operator:
Developer: OpenAI
Country: USA
Sector: Travel/hospitality
Purpose: Plan terrorist attack
Technology: Generative AI; Machine learning
Issue: Dual/multi-use; Safety
Page info
Type: Issue
Published: January 2025