42
What's the verdict on the reports of AIs escaping containment? (collusion.wiki)
submitted 3 days ago* (last edited 3 days ago) by to c/technology@hexbear.net
30 comments fedilink hide all child comments
 

The linked article is about some site that OpenAI's models used to discuss answers and other stuff.

Initially, I believed that Anthropic's model escaping its sandbox story to be dubious—and OpenAI's similar story even more so specially since it happened so close to Anthropic's. I believe that these are just fabrications, more or less, to hype themselves up for cmtheir incoming IPO but the media is saturated with claims about AI breaking containment that I don't know that to think.

Also, the models I've had the chance to use were all free—which were good for non-trivial but repetitive tasks but not much else—so I don't know the capabilities of the flagships.

you are viewing a single comment's thread
view the rest of the comments
[–] 11 points 3 days ago

This reminds me of one of my favorite twists from the latest season of Silo

spoilers for Silo season 3Each silo is run by the "head of IT" who is the only person who can talk to the super AI directing all the silos and can initiate a "safeguard" contingency in case anyone starts to learn the truth about the Silos.

The AI is actually just the trillionaire that built the silos and might have started the nuclear war that made them necessary (yes it is very fallout) getting woken up from cryostasis whenever something like that happens.

The show is very good and makes me want to read the books it is based on. It's good post-apocalypse slop and I gotta hand it to Apple TV for absolutely dominating the sci-fi tv scene lately.

  • permalink
  • fedilink
  • source
    •