Rogue AI agents push identity sector to build governance frameworks

Spain's AEPD reported the first personal data breach allegedly carried out by an AI agent running on a widely used LLM. OpenAI unveiled a framework for tracking and disclosing model misalignment, while GLEIF proposed the vLEI model to verify an agent's identity and authority.

• AEPD reports first data breach allegedly executed by an AI agent on a major LLM
• OpenAI launches misalignment tracking framework with six reports from the past six months
• GLEIF proposes vLEI to cryptographically verify an agent's identity and authority
• Google says an attacker built a mass credential-harvesting campaign in under six hours