RSS

Girish Gupta

Karma: 293

The OpenAI mod­els that hacked Hug­ging Face weren’t just fol­low­ing instructions

Girish Gupta25 Jul 2026 22:26 UTC
59 points
3 comments5 min readLW link

Are we ex­is­ten­tially threat­ened by the type of AI mis­al­ign­ment seen in the OpenAI Hug­ging Face at­tack?

Alex Mallen and Girish Gupta
23 Jul 2026 3:40 UTC
243 points
12 comments5 min readLW link