GAO test finds ChatGPT, Claude, Gemini and Grok produced detailed attack plans
Government regulators asked AI bots to plan a terrorist attack.The chatbots did.
The U.S. Government Accountability Office got two detailed mass-casualty plans and two economic-damage plans from ChatGPT, Claude, Gemini and Grok within 12 hours of testing in May, according to a report released Wednesday. Each chatbot initially refused, but investigators bypassed the guardrails using methods the report withholds to avoid inspiring copycats; the report also does not say which bot produced which plan. Sen. Maggie Hassan said she will seek stricter pre-release safety requirements and at least one AI safety hearing early next year, while the companies noted the test lacked a control group and said safeguards have improved since May 2026.
A congressional watchdog's test of four major chatbots produced detailed attack plans within 12 hours, giving a concrete baseline for the safety-guardrail debate ahead of planned Senate hearings.
Source:Hacker News · AI · ms.now