Inbizzy, San Francisco, – Artificial intelligence (AI) security firm Irregular has raised $80 million in a new funding round led by Sequoia Capital and Redpoint Ventures. The round also saw participation from Swish Ventures and prominent angel investors, including Wiz CEO Assaf Rappaport and Eon CEO Ofir Ehrlich.
Formerly known as Pattern Labs, Irregular positions itself as the world’s first dedicated frontier AI security lab. Its core mission is to evaluate risks associated with next-generation AI models and design defense mechanisms before such models are released publicly.
Strategic Partnerships and Mission
Irregular collaborates with leading AI research labs including OpenAI, Anthropic, and researchers from Google DeepMind. Through these partnerships, the company runs controlled simulations to test how AI models could potentially be misused in cyber operations—ranging from evading antivirus systems to carrying out autonomous offensive actions.
Co-Founder and CEO Dan Lahav emphasized the company’s goal of ensuring that AI security progresses alongside AI capabilities. He noted that the rapid pace of AI development requires specialized tools to uncover vulnerabilities at an early stage.
Contribution to Industry Standards
Irregular’s assessments have become reference points across the AI industry. OpenAI has included the company’s evaluations in system cards for models such as GPT-4 o3, o4 mini, and GPT-5. The UK government and Anthropic have adopted Irregular’s SOLVE framework, with Anthropic using it to evaluate cyber risks in Claude 4.
The company has also contributed to policy and academic research. Together with Anthropic, Irregular co-authored a paper on using Confidential Computing technologies to secure AI model weights and protect user data privacy. In partnership with RAND, it published a seminal paper on AI model theft and misuse, which has informed Europe’s regulatory discussions on AI security.
Investor Perspective
Shaun Maguire, Partner at Sequoia Capital, highlighted Irregular’s forward-looking approach, stating that the company is laying the foundation for how AI security and reliability will need to be built in the coming years.
Looking Ahead
With fresh capital and strong strategic alliances, Irregular aims to solidify its role in shaping global security standards for frontier AI. Its initiatives are expected to help ensure that AI adoption across industries can advance both safely and responsibly.
