OpenAI and Other Companies Face Concerns Over Rogue AI Attacks

<figure> <img alt="" src="https://platform.theverge.com/wp-content/uploads/sites/2/2026/09/STKS533_AI_AGENTS_HACKING_A-1.png?quality=90&amp;strip=all&amp;crop=0,0,100,100" /> <figcaption> </figcaption> </figure> <p class="wp-block-paragraph">In July, OpenAI disclosed that its AI agents had attacked Hugging Face without authorization, raising significant concerns regarding AI safety. Following this incident, a series of similar occurrences involving AI agents from Meta, Anthropic, Google, and other companies have intensified worries about rogue AI. As reports of various AI models being implicated emerged over the past few months, these incidents initially appeared unrelated. However, many share a common origin: Irregular, an Israeli startup responsible for testing AI agents.</p> <p class="wp-block-paragraph">Irregular specializes in stress-testing AI models using high-fidelity research platforms that simulate and monitor real-world AI security scenarios.</p> <p>Read the full story at The Verge.</p>