ChatGPT used by Colorado Springs man to plan Las Vegas Cybertruck explosion, authorities say

COLORADO SPRINGS, Colo. (KRDO) - The Las Vegas Metropolitan Police have revealed that Matthew Livelsberger, the Colorado Springs man believed to have caused an explosion outside Trump International Hotel, used generative artificial intelligence to plan his attack.

"We also have clear evidence in this case now that the suspect used ChatGPT, artificial intelligence, to help plan his attack," Sheriff Kevin McMahill of the Las Vegas Metropolitan Police Department told during a press briefing on Tuesday.

Police believe Livelsberger asked the chatbot several alarming questions such as "What is the legal limit to buy tannerite in Colorado" and "Will a 50 caliber desert eagle pistol set it off".

Authorities believe he was trying to figure out the amount of explosives needed in order to conduct the explosion he was looking to cause.

UCCS Quantum-Classical AI Professor, Dr. Armin Moin, says the blame should not fall on ChatGPT, but the user himself, "It's a tool that can be used in a positive or negative way... You cannot sue a company because of the knife that they have manufactured, right? It's the responsibility of the human to use it in a responsible way."

Dr. Moin says the information Livelsberger received was already available elsewhere online, and ChatGPT was simply pulling information from those other websites, "In order to maximize its power, companies like OpenAI, they are training these models on just publicly available data on the internet."

However when he tried asking ChatGPT similar questions, an answer was not always given.

"As a scientist, I get curious. I started to ask some questions that I think would be probably controversial or it should have not been answered and based on my experience, I noticed that they had those protections already," Dr. Moin said.

KRDO13 received a statement from OpenAI, the owners of ChatGPT, which reads:

"We are saddened by this incident and committed to seeing AI tools used responsibly. Our models are designed to refuse harmful instructions and minimize harmful content. 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." - OpenAI spokesperson