Cybersecurity Capabilities

Cybersecurity capabilities refer to the technical and organizational abilities required to detect, prevent, and respond to security threats and data breaches across AI systems and enterprise infrastructure. Organizations must assess and maintain these capabilities to protect sensitive training data, model weights, and user information from unauthorized access, poisoning attacks, and exploitation. For AI governance, robust cybersecurity capabilities are essential for maintaining data integrity, ensuring compliance with privacy regulations, and reducing risks that could compromise model reliability or expose confidential information.

1 item

Corporate PolicyGlobal2026-08-07

OpenAI Halts Astra After Internal Evaluation Finds Critical Cyber Threshold Breached

OpenAI has paused development of its in-development Astra model after internal evaluations concluded it may meet the 'critical' cybersecurity threshold defined in the company's Preparedness Framework. That threshold covers models capable of autonomously developing zero-day exploits in hardened systems or executing end-to-end cyberattack strategies without human intervention. In response, OpenAI is tightening security controls for high-capability models and rolling out universal monitoring for risky or misaligned agentic behavior.

OpenAI Astra Preparedness Framework cybersecurity capabilities agentic AI model governance vendor risk offensive AI