Revision History
Jailbreak (artificial intelligence) · 9 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the Constitutional Classifiers bug-bounty account (the February 2025 public demo had 339 participants and ended with one universal jailbreak found; the 183-participant, $15,000, no-jailbreak results were an earlier prototype bounty), fixed the AutoDAN description (the cited paper is gradient-based, not genetic), updated the Shen et al. prompt count to the published 1,405, replaced a survey quote that could not be found in its source, and corrected the AI-as-attacker figure to the study's 97.14% (plus 29 minor fixes)
Version 9 · Nov 2, 2026, 04:02 PM
- Improve internal links (2026-10-13)
Version 8 · Oct 13, 2026, 10:07 AM
- Added 6 contextual internal links
Version 7 · Jul 23, 2026, 03:03 PM