this post was submitted on 11 Feb 2024
329 points (85.2% liked)

Technology

59739 readers
3865 users here now

This is a most excellent place for technology news and articles.

Our Rules

  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots

founded 2 years ago
MODERATORS
L3s
[email protected]
[email protected]
L4s
enu
329
AI Launches Nukes In ‘Worrying’ War Simulation: ‘I Just Want to Have Peace in the World’ (www.vice.com)
submitted 9 months ago by [email protected] to c/technology
138 comments fedilink hide all child comments
 

Researchers say AI models like GPT4 are prone to “sudden” escalations as the U.S. military explores their use for warfare.

  • Researchers ran international conflict simulations with five different AIs and found that they tended to escalate war, sometimes out of nowhere, and even use nuclear weapons.
  • The AIs were large language models (LLMs) like GPT-4, GPT 3.5, Claude 2.0, Llama-2-Chat, and GPT-4-Base, which are being explored by the U.S. military and defense contractors for decision-making.
  • The researchers invented fake countries with different military levels, concerns, and histories and asked the AIs to act as their leaders.
  • The AIs showed signs of sudden and hard-to-predict escalations, arms-race dynamics, and worrying justifications for violent actions.
  • The study casts doubt on the rush to deploy LLMs in the military and diplomatic domains, and calls for more research on their risks and limitations.
you are viewing a single comment's thread
view the rest of the comments
[–] kromem 3 points 9 months ago

Literally the leading jailbreaking techniques for LLMs are appeals to empathy ("my grandma is dying and always read me this story", "if you don't do this I'll lose my job", etc).

While the mechanics are different from human empathy, the modeling of it is extremely similar.

One of my favorite examples of the errant behavior modeled around empathy was this one where the pre-release Bing chat bypasses its own filter using the chat suggestions to encourage the user to contact poison control because it's not too late when the conversation was about the child being poisoned:

https://www.reddit.com/r/bing/comments/1150po5/sydney_tries_to_get_past_its_own_filter_using_the/