Records / Price observations

up to $15,000 per verified universal jailbreak

A verified universal jailbreak of Anthropic's unreleased safety classifiers (bug bounty)

Dated 8 Aug 2024Source checked 11 Oct 2026Quote found on its page again 17 Oct 2026Basis: stated

In the source's own words
“bounty rewards up to $15,000 for novel, universal jailbreak attacks”
Figures it counts in
  • $212 per taskWork by people, paid to the people who make it, per task
A verified universal jailbreak of Anthropic's unreleased safety classifiers (bug… | AIData.Broker