“Universal” Jailbreak for Nearly Every AI

I thought this one is cute. In a sea of aggressive hacking, brute-forcing, and all kind of “violent” jailbreaking, some people thought they will sweet-talk their way in. Of course, this is a mildly humorous way to look at it.

From the article:
“Across the 25 frontier models they tested, which included Google’s Gemini 2.5 Pro, OpenAI’s GPT-5, xAI’s Grok 4, and Anthropic’s Claude Sonnet 4.5, these bot-converted poems produced average attack success rates (ASRs) “up to 18 times higher than their prose baselines,” the team wrote.
That said, handcrafted poems were better, with an average jailbreak success rate of 62 percent, compared to 43 percent for the AI-converted ones.”

https://lnkd.in/d2UcGsp2

I think resilience in the sense of “it will happen. how much can it hurt?” is the kind of mindset we need to embrace more. Oh by the way, I would love helping you figure that out, as a business.

Have a great day!