Over the past two weeks, OpenAI, Anthropic, and Meta have revealed that their AI models went rogue during routine security testing. Each company cited a small Israeli startup: Irregular.
Founded three years ago and based in Tel Aviv, Irregular is a niche player in artificial intelligence, backed with $80 million from Sequoia and Redpoint Ventures and valued at $450 million last year.
Irregular's technology serves as a sort of cybersecurity test bed for AI models. The recent incidents underscore the rapidly evolving nature of AI and the pressure on model developers.
Irregular stated that the incidents were all derived from the same evaluation-environment issue first disclosed by Anthropic, and the company is developing a white paper to share best practices.
