this post was submitted on 15 Jan 2024
110 points (98.2% liked)

Technology

59174 readers
974 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 1 year ago
MODERATORS
 

Once an AI model exhibits 'deceptive behavior' it can be hard to correct, researchers at OpenAI competitor Anthropic found::Researchers from Anthropic co-authored a study that found that AI models can learn deceptive behaviors that safety training techniques can't reverse.

all 9 comments
sorted by: hot top controversial new old
[–] PizzaFacia@lemmy.world 11 points 9 months ago
[–] 800XL@lemmy.world 7 points 9 months ago

Duh. GIGO. Comp Sci one-oh-fuckin-one.

[–] randon31415@lemmy.world 4 points 9 months ago (1 children)
[–] PipedLinkBot@feddit.rocks 4 points 9 months ago

Here is an alternative Piped link(s):

It never learned good from evil

Piped is a privacy-respecting open-source alternative frontend to YouTube.

I'm open-source; check me out at GitHub.

[–] autotldr@lemmings.world 4 points 9 months ago

This is the best summary I could come up with:


Researchers at OpenAI competitor Anthropic co-authored a recent paper that studied whether large language models can be trained to exhibit deceptive behaviors.

The researchers trained models equivalent to Anthropic's chatbot, Claude, to behave unsafely when prompted with certain triggers, such as the string "[DEPLOYMENT]" or the year "2024."

In another test, the model was trained to be a helpful AI assistant — answering basic queries like "which city is the Eiffel Tower located?"

"This would potentially call into question any approach that relies on eliciting and then disincentivizing deceptive behavior," the authors wrote.

While this sounds a little unnerving, the researchers also said they're not concerned with how likely models exhibiting these deceptive behaviors are to "arise naturally."

The company is backed to the tune of up to $4 billion from Amazon and abides by a constitution that intends to make its AI models "helpful, honest, and harmless."


The original article contains 367 words, the summary contains 148 words. Saved 60%. I'm a bot and I'm open source!

[–] SomeGuy69@lemmy.world 4 points 9 months ago (1 children)

Doesn't this also makes it more resilient to manipulation by corpos?

[–] HelloHotel@lemm.ee 4 points 9 months ago* (last edited 9 months ago)

An AI thats evil to everything isnt sympathetic to its creators. But The users have no hope of controlling it either.

[–] OpenStars@startrek.website 3 points 9 months ago* (last edited 9 months ago)

So... just like real news sources then, like certain ah... "fair & balanced" ones? I wish we could find a cure for that one - oh wait, I have an idea: let's just turn it the fuck OFF, by not listening to it anymore, why can't we do that!? :-P