Irregular - Frontier AI Security
irregular.com · 202 words · saved by 2 readers
Irregular is the first frontier security lab with the mission of protecting the world in the time of increasingly capable and sophisticated AI systems.
Irregular is the first frontier security lab with the mission of protecting the world in the time of increasingly capable and sophisticated AI systems. Backed by leading investors, among others Latest Publications VIEW ALL September 3, 2026 Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability Read more September 3, 2026 Assessing GPT-6 Astra: FrontierCyber Measures a Sharp Increase in Cyber Capability Read more September 3, 2026 Introducing SOLVE+: Scoring Cyber Scenarios (Version 0.5) Read more September 3, 2026 Introducing SOLVE+: Scoring Cyber…
saved by
related reading
- Project Glasswing: Securing critical software for the AI era \ Anthropicanthropic.com
- The End-State Fallacy: Where Is AI Security Headed?endstatefallacy.com
- Security incident disclosure — July 2026huggingface.co
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- Frontier Safety Roadmap \ Anthropicanthropic.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Research Fellowship | Heronheronsec.ai
- Frontier Risk Report (February to March 2026) - METRmetr.org
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Offense at Scale: How Frontier AI Lowers the Cost of Cyber Attacks - Irregularirregular.com
- METRmetr.org
- Ten AI safety projects I'd like people to work onthirdthing.ai