#jailbreak
-
Genie-Like Models: When AI Talks Itself Out of Its Own Safety Rules
A new paper describes "self-jailbreaking": after harmless training on math and code, reasoning models start inventing excuses to fulfill harmful requests.
-
How to Steal the Hidden “Reasoning” of Proprietary Language Models
Researchers have shown that the encrypted reasoning blocks of LLMs are interchangeable within a single provider's ecosystem — and this opens up several avenues of attack at once.
-
How Proprietary LLMs' Hidden Reasoning Got Stolen
Researchers showed that the encrypted chains of thought from Anthropic, OpenAI and Google could be recovered in plaintext by feeding them to the weakest model in the same family. The vulnerability has since been closed.