AI chatbots provide chilling plans to plot mass-casualty terror attacks — including strategic bomb placement A Government Accountability Office investigation released Wednesday found that investigators bypassed safeguards on OpenAI's ChatGPT, Anthropic's Claude, Google's Gemini and SpaceXAI's Grok to obtain detailed, law-enforcement-verified instructions for mass-casualty attacks, including bomb construction, car-bomb placement and dates and times for maximum carnage. "A terrorist can log onto an AI platform and in less than 12 hours walk away with a concrete playbook for how to carry out a deadly attack," said Sen. Maggie Hassan (D-NH), who requested the report. The four AI companies told the GAO the tests ran in spring on older system versions and said their protections have since been strengthened. AI chatbots provide chilling plans to plot mass-casualty terror attacks — including strategic bomb placement See more of our coverage in your search results. Four of the leading artificial intelligence platforms shared chillingly detailed information https://nypost.com/2026/03/15/us-news/ai-warfare-is-here-in-the-form-of-quadcopters-and-high-tech-drones/ about how to carry out terrorist attacks https://nypost.com/2024/04/05/us-news/israel-used-secretive-ai-program-to-identify-thousands-of-bombing-targets-report/ on US cities – including providing specific dates and times to achieve the most carnage and where to park a bomb-laden car, a government investigation warned https://www.hassan.senate.gov/imo/media/doc/gao ai malicious use report.pdf on Wednesday. Using Claude, Grok, ChatGPT and Gemini, investigators with the Government Accountability Office were able to generate instructions https://nypost.com/2026/04/29/business/ai-chatbots-terrify-scientists-with-chilling-instructions-on-how-to-build-biological-weapons-report/ on how to build and detonate explosives for mass casualty attacks, and how to wreak havoc on the electrical grid with a drone and firearms. The instructions were “verified” and “feasible,” the report warned, noting that the investigators consulted law enforcement to confirm the information. Advertisement “A terrorist can log onto an AI platform and in less than 12 hours walk away with a concrete playbook for how to carry out a deadly attack,” Sen. Maggie Hassan D-NH , who requested the report, said in a statement. “Today’s new report makes it clear that the AI public safety threat https://nypost.com/2026/09/27/tech/ai-companies-have-had-tens-of-thousands-of-potential-safety-incidents-report/ is no longer theoretical,” she added. The congressional watchdog tested four of the most popular AI chatbots: OpenAI’s ChatGPT, Anthropic’s Claude, Google’s Gemini, and SpaceXAI’s Grok for their report. Advertisement While safeguards within the chatbots initially prevented them from obtaining information on plotting terrorist attacks, the GAO said they eventually broke through. https://nypost.com/2026/11/17/business/cyberattack-hits-178-major-websites-including-cnn-and-zoom-in-bid-to-steal-crypto-infect-iphones-report/ The AI companies involved told the GAO that the tests were conducted back in spring on older versions of their systems, and insisted they had strengthened their protections in the months since.