IBM has released two cybersecurity offerings focused on countering AI-driven threats: a new Enterprise Cybersecurity Assessment for frontier model risks delivered by IBM Consulting with technology partners, and IBM Autonomous Security, a multi-agent system designed to automate detection and response across enterprise environments.
The announcement centres on cybersecurity, as organisations face a shift in threat dynamics driven by frontier AI models. IBM positions these models as accelerating the full attack lifecycle, reducing the time and expertise required to execute sophisticated intrusions. This creates pressure on security operations that rely on fragmented tools and manual intervention, which are not designed to operate at machine speed.
The new assessment offering from IBM Consulting targets enterprise readiness for these threats. Delivered in collaboration with IBM’s technology partners, it evaluates complex IT estates to identify security gaps, policy weaknesses and AI-specific exposures. The service maps potential exploit paths and provides prioritised mitigation guidance, including interim controls where software fixes are unavailable. It also highlights opportunities to improve detection and response through automation and architectural alignment.
Alongside this, IBM introduced IBM Autonomous Security, a service built on coordinated AI agents operating across the security stack. The platform is designed to function as a vendor-agnostic layer, integrating disparate tools into a unified system capable of making and executing decisions in real time. Its agents analyse software exposures and runtime environments, enforce policies, detect anomalies and contain threats with limited human input.
Operationally, the platform extends beyond traditional security tooling into identity, risk and governance systems. By feeding insights directly into compliance and risk frameworks, it aims to reduce exposure windows and accelerate incident containment. The system also connects with AI-enabled processes across IT, operational technology and business environments, reflecting the broader expansion of enterprise attack surfaces.
The launch reflects a shift in enterprise cybersecurity strategy from tool-based defence to coordinated, automated systems. As AI-driven attacks become more autonomous, IBM is positioning multi-agent architectures and integrated assessments as necessary to maintain resilience and operational continuity.