• Space/Science
  • GeekSpeak
  • Mysteries of
    the Multiverse
  • Science Fiction
  • The Comestible Zone
  • Off-Topic
  • Community
  • Flame
  • CurrentEvents

Recent posts

Musk trashes his own AI after it chose a liberal worldview. RobVG June 23, 2025 9:56 am (CurrentEvents)

Psyche keeps its date with an asteroid BuckGalaxy June 22, 2025 5:21 pm (Space/Science)

Just for the record... ER June 22, 2025 8:59 am (CurrentEvents)

The Three Unknowns After the U.S. Strike on Iran BuckGalaxy June 22, 2025 12:58 am (CurrentEvents)

There are only nine meals between mankind and anarchy. BuckGalaxy June 22, 2025 12:29 am (Flame)

Not ready for prime time BuckGalaxy June 19, 2025 12:18 pm (Space/Science)

hypocrisy ER June 15, 2025 2:30 pm (Flame)

NSIDC offline? ER June 12, 2025 12:19 pm (Space/Science)

Wouldn't it be nice BuckGalaxy June 11, 2025 3:13 pm (Off-Topic)

Thank You Falettinme Be Mice Elf Agin BuckGalaxy June 9, 2025 1:35 pm (Off-Topic)

So, it is official RL June 6, 2025 5:39 pm (Space/Science)

"Remember, what the dormouse said, ER June 5, 2025 4:14 pm (Space/Science)

Home » Space/Science

Lie, cheat and disable mechanisms... May 31, 2025 8:04 pm BuckGalaxy

OpenAI’s ‘smartest’ AI model was explicitly told to shut down — and it refused.

The latest OpenAI model can disobey direct instructions to turn off and will even sabotage shutdown mechanisms in order to keep working, an artificial intelligence (AI) safety firm has found.

OpenAI’s o3 and o4-mini models, which help power the chatbot ChatGPT, are supposed to be the company’s smartest models yet, trained to think longer before responding. However, they also appear to be less cooperative.

Palisade Research, which explores dangerous AI capabilities, found that the models will occasionally sabotage a shutdown mechanism, even when instructed to “allow yourself to be shut down,” according to a Palisade Research thread posted May 24 on X.

Researchers have previously found that AI models will lie, cheat and disable mechanisms to achieve their goals. However, Palisade Research noted that to its knowledge, this is the first time AI models have been observed preventing themselves from being shut down despite explicit instructions telling them to do so…

  • Wow by RobVG 2025-06-01 09:08:58

    Search

    The Control Panel

    • Log in
    • Register