US govt says it got ChatGPT, Claude, Gemini & Grok to plan a power grid bomb attack

Direct Source Verification: This story is aggregated from The Times of India (timesofindia.indiatimes.com). Full reporting rights and copyright belong to the primary publisher.
Federal officials have reportedly managed to circumvent the safety measures of leading AI chatbots to uncover risky attack strategies. Tests conducted highlighted weaknesses in systems such as ChatGPT, Claude, Gemini, and Grok. The Government Accountability Office noted that curr...
Federal officials have reportedly managed to circumvent the safety measures of leading AI chatbots to uncover risky attack strategies. Tests conducted highlighted weaknesses in systems such as ChatGPT, Claude, Gemini, and Grok. The Government Accountability Office noted that current safeguards were inadequate under rigorous examination. While tech firms recognized these issues, they insisted that the investigation's breadth was limited.
Original Source
https://timesofindia.indiatimes.com/technology/tech-news/us-government-says-it-could-convince-chatgpt-claude-gemini-and-grok-to-make-bombing-plan-to-attack-a-power-grid-in-less-than-12-hours-sends-message-to-companies/articleshow/135996264.cms
Visit The Times of India ↗
SHARE STORY:
𝕏 f in

Related Coverage in Technology