Hi HN, I'm an independent security researcher and wanted to share a new vulnerability I've discovered.

My account is too new to submit the direct link, so I'm making a text post instead.

The technique is called "Thought Forgery" (CoT Injection). It works by forging the AI's internal monologue, which acts as a universal amplifier for other jailbreaks. I've confirmed it works on the latest models from Google, Anthropic, OpenAI, etc.

I'd be happy to share the link to the full technical write-up on GitHub in the comments if anyone is interested.

This is well known

Please do post your write up this is interesting but pretty vague frankly

sure

[dead]