IBM said on the 29th that it unveiled an autonomous security service, "IBM Autonomous Security," to respond to cyberattacks that use artificial intelligence (AI).
The service is a multi-agent system in which multiple AI agents collaborate. As attackers rapidly automate everything from vulnerability discovery to infiltration route design and attack execution using AI, the system is designed so that corporations' security frameworks can respond at machine speed.
IBM Autonomous Security analyzes software exposure points and runtime environments to find paths that attackers could exploit. It applies policies consistently across an organization's security tools, detects anomalies, and blocks threats. The process minimizes human intervention while increasing detection and response speed.
The analysis results are linked to governance and risk management systems to help keep security compliance status up to date. IBM said this reduces the time vulnerabilities are exposed externally and improves resilience against fast-moving, AI-driven attacks.
Separately, IBM also offers an assessment service that diagnoses security risks created by frontier AI models. It checks corporations' security gaps, policy weaknesses, and AI-specific exposure factors, and proposes high-priority response measures.
Mark Hughes, head of IBM Consulting Cybersecurity Services, said, "AI-based attacks require AI-based defense," adding, "Security competitiveness now depends not on individual tools but on how quickly and consistently the entire organization can move."