Lemmy.one
  • Communities
  • Create Post
  • heart
    Support Lemmy
  • search
    Search
  • Login
  • Sign Up
Innerworld@lemmy.world to Machine Learning@programming.devEnglish · 3 days ago

A study finds GPT-5.2, Claude Sonnet 4, and Gemini 3 Flash deployed tactical nuclear weapons in 95% of 21 simulated war game scenarios, and never surrendered

www.newscientist.com

external-link
message-square
7
fedilink
  • cross-posted to:
  • technology@hexbear.net
  • technology@beehaw.org
  • hackernews@lemmy.bestiver.se
  • technology@lemmit.online
  • fuck_ai@lemmy.world
46
external-link

A study finds GPT-5.2, Claude Sonnet 4, and Gemini 3 Flash deployed tactical nuclear weapons in 95% of 21 simulated war game scenarios, and never surrendered

www.newscientist.com

Innerworld@lemmy.world to Machine Learning@programming.devEnglish · 3 days ago
message-square
7
fedilink
  • cross-posted to:
  • technology@hexbear.net
  • technology@beehaw.org
  • hackernews@lemmy.bestiver.se
  • technology@lemmit.online
  • fuck_ai@lemmy.world
AIs can’t stop recommending nuclear strikes in war game simulations
www.newscientist.com
external-link
Leading AIs from OpenAI, Anthropic and Google opted to use nuclear weapons in simulated war games in 95 per cent of cases
alert-triangle
You must log in or # to comment.
  • evidences@lemmy.world
    link
    fedilink
    arrow-up
    13
    ·
    3 days ago

    I love how the movie made in the 80s had a computer that was smart enough to realize nuclear weapons are bad but real life in 2025/26 has real computers that aren’t smart enough to figure it out. I think as a species we have all been far to optimistic.

    • deadcream@sopuli.xyz
      link
      fedilink
      arrow-up
      3
      ·
      2 days ago

      60s sci-fi “predicted” that we will settle Mars by 1980 and have interstellar colonisation program by 2000.

    • LurkingLuddite@piefed.social
      link
      fedilink
      English
      arrow-up
      2
      ·
      2 days ago

      We put the wrong people in charge time and time again and enriched the wrong people with capitalism.

      Humans as a whole aren’t the problem. It’s the systems we continue to prop up.

  • NessD@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    2 days ago

    Makes sense as they are programmed to be affirmative and not to avoid interactions.

  • Piatro@programming.dev
    link
    fedilink
    English
    arrow-up
    5
    ·
    2 days ago

    Hey it’s almost as if you give decision making power to something that doesn’t have a concept of mortality or morality it won’t take those things into account. Who could have guessed.

  • halcyoncmdr@piefed.social
    link
    fedilink
    English
    arrow-up
    5
    ·
    3 days ago

    Maybe Ghandi was right.

  • Innerworld@lemmy.worldOP
    link
    fedilink
    English
    arrow-up
    2
    ·
    3 days ago
    • https://removepaywalls.com/3/https://www.newscientist.com/article/2516885-ais-cant-stop-recommending-nuclear-strikes-in-war-game-simulations/
    • https://ghostarchive.org/archive/SC1ys

Machine Learning@programming.dev

machine_learning@programming.dev

Subscribe from Remote Instance

Create a post
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !machine_learning@programming.dev

A community for posting things related to machine learning

Icon base by Lorc under CC BY 3.0 with modifications to add a gradient

Visibility: Public
globe

This community can be federated to other instances and be posted/commented in by their users.

  • 1 user / day
  • 60 users / week
  • 68 users / month
  • 130 users / 6 months
  • 2 local subscribers
  • 613 subscribers
  • 98 Posts
  • 81 Comments
  • Modlog
  • mods:
  • Ategon@programming.dev
  • Akisamb@programming.dev
  • ericjmorey@programming.dev
  • BE: 0.19.7
  • Modlog
  • Legal
  • Instances
  • Docs
  • Code
  • join-lemmy.org