AI chatbots tend to choose violence and nuclear strikes in wargames
AI chatbots tend to choose violence and nuclear strikes in wargames
www.newscientist.com AI chatbots tend to choose violence and nuclear strikes in wargames
As the US military begins integrating AI technology, simulated wargames show how chatbots behave unpredictably and risk nuclear escalation
4
crossposts
You're viewing a single thread.
View all comments
54
comments
So do humans.
38 1 ReplyHere's a wild thought. Maybe that's why the chat bot (I assume LLM) does it too, because it's been trained on us! 🤯
27 0 ReplyI learned it from watching you!
2 0 Reply
Where are all these nuclear strikes?
5 1 ReplySid Meier's Civilization games
9 0 ReplyGhandi has the right idea.
1 0 Reply
You've viewed 54 comments.
Scroll to top