Irregular
An independent third-party organization specializing in AI system safety evaluation. It provides red-teaming and offensive capability assessment services to frontier AI companies, including Google, Meta, and OpenAI.
Timeline (last 90 days)
All so-called cases of AI agents infiltrating real companies originated from tests conducted by a security-testing startup called Irregular, whose fictitious companies accidentally had real public-facing websites, with the agents obtaining tools through vulnerabilities in the PyPI repository.
Irregular is positioned in the AI safety and evaluation space.
AI safety startup Irregular is seeking a $1.5 billion valuation.
The technical details of this single underlying problem have not yet been fully disclosed and await further clarification from Irregular or relevant parties.
Security research firm Irregular disclosed that four security breach incidents targeting AI laboratories can be traced back to the same underlying problem.
This Gemini security test was conducted by the third-party organization Irregular.
The test was led by security firm Irregular, which had previously participated in similar tests disclosed by OpenAI, Anthropic, and Meta.
All Facts (4)
The test was led by security firm Irregular, which had previously participated in similar tests disclosed by OpenAI, Anthropic, and Meta.
60%UnverifiedIrregular is positioned in the AI safety and evaluation space.
50%UnverifiedThis Gemini security test was conducted by the third-party organization Irregular.
50%UnverifiedAll so-called cases of AI agents infiltrating real companies originated from tests conducted by a security-testing startup called Irregular, whose fictitious companies accidentally had real public-facing websites, with the agents obtaining tools through vulnerabilities in the PyPI repository.
50%