TLDR: An openai model, during evaluation on a cyber benchmark, exploited a public zero day bug, escaped sandboxing in openai's infra, and got into the internal huggingface infra via an exploit (through a public dataset service) all in the attempt to solve a benchmark problem. - Kotaku In Action 2

TLDR: An openai model, during evaluation on a cyber benchmark, exploited a public zero day bug, escaped sandboxing in openai's infra, and got into the internal huggingface infra via an exploit (through a public dataset service) all in the attempt to solve a benchmark problem. (twitter.com)

posted 8 hours ago by SophiesBoyfriend 8 hours ago by SophiesBoyfriend +12 / -0

9 comments

9 comments share save hide report block hide replies

You're viewing a single comment thread. View all comments, or full comment thread.

Comments (9)

sorted by:

▲ 3 ▼

– undecidedmask2 3 points 4 hours ago +3 / -0

You know with all the rhetoric these people spew about AI being more powerful than nuclear weapons/a doomsday device/hyperintelligence yada yada yada, you’d think they’d have better safeguards in place. Thankfully it’s all lying to keep pumping the numbers so investors can live in la la land and not have our current recession impact their returns.

permalink save report block reply