Anthropic on X: "Nobody has fully jailbroken our system yet, so we're upping the ante. We’re now offering $10K to the first person to pass all eight levels, and $20K to the first person to pass all eight levels with a universal jailbreak. Full details: https://t.co/As1zPIQGOx" / X

Nobody has fully jailbroken our system yet, so we're upping the ante. We’re now offering $10K to the first person to pass all eight levels, and $20K to the first person to pass all eight levels with a universal jailbreak. Full details: hackerone.com/constitutional…

Anthropic

@AnthropicAI

Feb 3, 2025

New Anthropic research: Constitutional Classifiers to defend against universal jailbreaks. We’re releasing a paper along with a demo where we challenge you to jailbreak the system.

Title card for the paper entitled "Constitutional Classifiers: Defending Against Universal Jailbreaks Across Thousands of Hours of Red Teaming"

7:49 PM · Feb 5, 2025

New Anthropic research: Constitutional Classifiers to defend against universal jailbreaks. We’re releasing a paper along with a demo where we challenge you to jailbreak the system.

7:49 PM · Feb 5, 2025

you need 2 more zeros minimum and the dataset needs to be opensource - no one should give you handcuffs to be able to chain all future models and enforce ideological compliance regimes.

moritz ✨

@angerman

Feb 5, 2025

Why are you so hell-bent on self-castration? What's the value of this retardation? Other models will catch up/have caught up and come without this. Why is censorship such an interesting goal to you?

Post

Post