AI-Assisted Bomb Making Arrests in NYC

Federal prosecutors have indicted a 55-year-old man from Long Island, New York, alleging that he built seven homemade bombs with assistance from artificial intelligence. According to court documents, Michael Gann described this process as “easier than buying gun powder.”

Gann reportedly transported the bombs to New York City, where he stored five of them, along with four shotgun shells, on the rooftop of an apartment building in the SoHo neighborhood. His intent was to combine the shotgun shells with one or more of the improvised explosive devices (IEDs).

On March 27, Gann posted a concerning message directed at President Donald Trump on his official X account, implying violent intentions. His post stated, “Dear @POTUS, I’m thinking just now here in NYC that it’s too bad that the wall wasn’t built before the National Guard would have to come here for the Protests and Riots.” Five weeks later, Gann was arrested following an Instagram post that read, “Who wants me to go out to play like no tomorrow?”

Investigations indicated that Gann had used two household compounds ordered online to manufacture the explosives, purchasing over 200 cardboard tubes and more than 50 feet of fuses. One device he fabricated contained approximately 30 grams of explosive powder, greatly exceeding the legal limit for consumer fireworks.

According to authorities, Gann acted independently and not as part of a group. Christopher Raia, head of the FBI’s New York field office, stated, “Gann allegedly produced multiple improvised explosive devices intended for use in Manhattan. Thanks to the successful collaboration between law enforcement agencies, Gann was swiftly brought to justice before he could harm innocent civilians.”

Witnesses reported alarming conversations with Gann, who described his mixing of explosives and expressed disdain for inaction regarding neighborhood problems, alluding to potential targets within his locality.

Upon his arrest, law enforcement agents observed Gann carrying a shoulder bag on a street. When identified as a suspect, he claimed he was en route to the fire department to surrender the devices.

After being read his rights, Gann disclosed that he had utilized AI to identify which chemicals to buy and mix for his explosive creations. Initially, he produced four devices, discarding three from the Williamsburg Bridge, leading to one that fell onto subway tracks and was subsequently recovered.

Interim U.S. Attorney Jay Clayton condemned Gann’s actions, stating, “Gann built explosive devices, stored them on a rooftop in SoHo, and threw one onto the subway tracks—putting countless lives at risk.”

This disturbing case raises questions about the use of AI in dangerous contexts and highlights the need for stronger oversight and security measures to prevent future incidents.