Post

They were like high-school students trying to hack into the textbook company to cheat on their final exam. Only these hackers weren’t human.

on.wsj.comHow the Futuristic Hack by Rogue OpenAI Models UnfoldedThe incident serves as an early example of loss-of-control scenarios long feared by artificial-intelligence safety researchers.

24/7/26, 12:17 am12 reposts21 likes

Paywall.

@oconnellmark.bsky.social

24 July

Evergreen Innovations LLC

@evergreeninv.bsky.social

24 July

AI works within the guardrails, guidelines and information which is programmed into it. The fact that it was able to do this means that it didn't have any guardrails against it. Maybe we should only have people with foresight and common sense exploring the frontier of this technology...