AIUC Raises $40 Million to Certify Enterprise AI Agent Security

AIUC has secured $40 million in Series A funding to expand its security certification standards for enterprise AI agents.

The San Francisco-based Artificial Intelligence Underwriting Company (AIUC) has brought its total funding to $55 million following the round. The investment was led by Ribbit Capital, with additional support from First Harmonic.

The AIUC-1 Security Framework

The company’s core offering, the AIUC-1 standard, provides a framework for evaluating the security and reliability of AI systems. The standard subjects models to approximately 5,000 adversarial risk scenarios and includes quarterly audits to address emerging threat vectors.

The testing process is designed to identify specific vulnerabilities, including prompt injections, jailbreaks, hallucinations, anomalous behaviour, and potential data leaks.

Addressing Enterprise Deployment Bottlenecks

AIUC founder Rune Kvist, who previously served as the first product hire at Anthropic, stated that many enterprises have approved AI agents for pilots, but deployment often stalls during the security review process. Kvist noted that providing clear evidence of security and reliability is now a critical requirement for enterprise adoption.

The AIUC-1 standard is already being used to certify several prominent AI agents and organisations, including UiPath, KPMG, ElevenLabs, Cursor, Harvey, Fin, and Lovable.

AIUC intends to use the new capital to expand its auditing capabilities and security standards to include frontier models.