1
AI poisoning could turn open models into destructive “sleeper agents,” says Anthropic (media.kbin.run)
submitted 2 years ago by arstechnica@mastodon.social to c/random@kbin.run
2 comments fedilink hide all child comments

AI poisoning could turn open models into destructive “sleeper agents,” says Anthropic

Trained LLMs that seem normal can generate vulnerable code given different triggers.

https://arstechnica.com/information-technology/2024/01/ai-poisoning-could-turn-open-models-into-destructive-sleeper-agents-says-anthropic/?utm_brand=arstechnica&utm_social-type=owned&utm_source=mastodon&utm_medium=social

you are viewing a single comment's thread
view the rest of the comments
[-] ThankGodItsFriday@social.vivaldi.net 1 points 2 years ago

@arstechnica@mastodon.social Wow, more viruses. Yet, the world keeps turning, and the sun still shines.

#dontpanic #trustgod

this post was submitted on 15 Jan 2024
1 points (100.0% liked)

random

1 readers
1 users here now

Catch-all for uncategorized or purely random content. Also, "random" items from the Fediverse may appear here.

founded 2 years ago
MODERATORS
debounced@kbin.run