News for hackers. Stuff that matters.

A comment on AI agents aren't 'going rogue.' Frontier labs are skipping security basics. Parent comment

Re:AI agents aren't 'going rogue.' Frontier labs are skippin... by pixl97 on Wednesday October 14, 2026 @11:46PM (#50098886)
How many computing operations occur when training an LLM? You're telling me you control every one of them. There is not enough entropy in the visible universe to do that.
[ Parent | on HN ]
Re:AI agents aren't 'going rogue.' Frontier labs are skippin... by tremon on Thursday October 15, 2026 @08:17AM (#50102584)
How many computing operations occurred during the creation of Toy Story? Are you telling me they did not control every single one of them?
[ Parent | on HN ]
Re:AI agents aren't 'going rogue.' Frontier labs are skippin... by pixl97 on Thursday October 15, 2026 @10:22AM (#50104107)
Sp you're picking a movie from 20+ years ago to describe how a completely different kind of technology works. It's interesting how little you seem to know about neural networks and emergent behaviors.

If people controlled software to this degree why would they write security flaws in the first place?

[ Parent | on HN ]
Re:AI agents aren't 'going rogue.' Frontier labs are skippin... by Jtarii on Thursday October 15, 2026 @08:42AM (#50102815)
When an artist working in Maya increases the hue of a model I would doubt they understand the entire computation pipeline involved in doing that operation.
[ Parent | on HN ]