Legal/Law/Criminal Justice and Reform

Florida Woman Nabbed After Chatbot Maker Flags Alleged Threat to ‘Shoot Up’ Sheriff’s Office

(Wikimedia Commons/Public/Tony Webster, CC BY-SA 4.0)

A chatbot conversation that an artificial intelligence giant reported to police has landed a Florida woman a felony charge.

Anthropic alerted law enforcement after Carli Michelle Heller of Bonita Springs, Florida, reportedly made threats against the Lee County Sheriff’s Office while using the company’s chatbot, leading to her arrest Wednesday, according to the office’s arrest report it sent to the Daily Caller News Foundation. As this case unfolds, lawmakers weigh how much responsibility AI companies should bear for policing their platforms, with Republican Sen. Josh Hawley of Missouri and Democratic Sen. Chris Murphy of Connecticut pushing a bill that could expose these tech giants to lawsuits.

“I’m going to shoot up the sheriff’s [the Lee County Sheriff’s Office] right the fuck now.,” Heller wrote in a Sept. 26 chat to Anthropic’s chatbot Claude, the Wednesday arrest report stated.

“THIS IS 100% LAST CHANCE I’M DONE. I GOT A NEW FUCKING GUN TODAY. FUCK YOU.” she said to the chatbot the next day.

She was identified by authorities and arrested at her home in Bonita Springs, according to the arrest report. She was charged with written threats of violence, a second-degree felony under Florida law.

Heller used AI as a “diary,” Sheriff Carmine Marceno told WINK News.

“Artificial intelligence is a powerful tool, and like any technology, it can be misused,” Sheriff Carmine Marceno said in a statement to the DCNF. “When someone uses AI to make or facilitate a threat, we have to take that threat seriously.”

“Users need to understand you are never truly anonymous, specifically AI chatrooms and searches,” Marceno added. “AI may be changing the way threats are created, but our commitment to protecting the people of Lee County has not changed.”

The Sheriff’s office told the DCNF that it does not track the number of tips received about AI.

Anthropic did not immediately respond to the DCNF’s request for comment.

Anthropic’s Safeguards Team checks how people use Claude to check for violations of the company’s policies, according to the company’s Usage Policy page. The company will slow or revoke a user’s access to its services for violations, or it may modify outputs in response to Usage Policy violations.

This is not the first case of a Claude conversation being turned over to police.

One chat about shooting an elementary school in San Antonio led the FBI to arrest a 22-year-old individual for third-degree terroristic threats, according to San Antonio Express-News. In another case, a user in San Francisco told the chatbot they had purchased an AR-15 and also threatened Anthropic CEO Dario Amodei, but the individual was not arrested or charged with a crime.