ChatGPT was used in the Las Vegas Cybertruck attack

Sunday 12th January 2025 04:41 AM

The explosion of a Tesla Cybertruck in front of the Trump hotel in Las Vegas on January 1 once again raises questions about the risks linked to artificial intelligence. According to the Las Vegas Metropolitan Police, the suspect, Matthew Livelsberger, a 37-year-old soldier, allegedly used ChatGPT to plan his act. This would be the first known case in the United States where an artificial intelligence model is involved in an attack of this nature.

ChatGPT at the heart of the preparations

According to the authorities, Livelsberger reportedly consulted ChatGPT for technical information related to explosivesprojectile speed, and the legality of fireworks in Arizona. This research, although based on publicly available data, would have helped overcome some obstacles in planning the explosion. The police also discovered a diary and a series of notes, which confirm that the attack was intended to send a “wake-up call” about the state of the country.

OpenAI defends its tool

Faced with these revelations, OpenAI, the company behind ChatGPT, published a press release. It says its model is designed to deny harmful instructions and regularly displays warnings about illegal activities based on queries. “We are deeply saddened by this incident and are working with law enforcement to support the investigation,” OpenAI said..

Except that critics already believe that these safeguards remain insufficient. Cybersecurity expert Paul Keener, for example, says that AI tools are still at an early stage of regulation. He even predicts: “Controls will strengthen as technology evolves.”

A turning point for public safety

Sheriff Kevin McMahill described this event as a “worrying moment”, insisting that the use of AI in this type of attack is clearly an unprecedented challenge for law enforcement. Moreover, details of the searches carried out via ChatGPT by the man behind the attack have been shared with other agencies to better anticipate possible similar scenarios.

The explosion caused minor injuries to seven people and little material damage. Livelsberger, who committed suicide at the scene, acted alone, according to police. His personal notes evoke political frustrationsbut no direct links with extremist groups have yet been established.

Even if this case remains isolated for the moment, it shows certain risks linked to misuse of artificial intelligence tools. The debates on the regulation of these technologies, already underway, could well intensify. While AI still has a lot to offer, it still poses new questions in terms of security and ethics.