Cohesity has rolled out Cohesity Agent Resilience, a new Cohesity Data Cloud capability engineered to discover, protect, and recover the infrastructure underlying enterprise AI agents.
Unveiled at Cohesity Catalyst, the new feature addresses a critical operational gap as automated workflows increasingly query databases, modify files, and use privileged credentials without direct human intervention. Industry projections indicate that up to 40% of enterprise applications will integrate task-specific AI agents by 2026, creating severe recovery challenges if misconfigurations, memory corruption, or malicious activities alter system states.
Cohesity Agent Resilience leverages the platform’s established snapshot architecture, immutable backups, and clean-room recovery mechanisms to safeguard agent memory and configuration settings. The capability delivers point-in-time recovery to restore agents to known-good states alongside precise restoration functionality for databases, file systems, and other downstream services impacted by automated agent actions. At launch, the system integrates natively with Amazon Bedrock, with platform support for Microsoft and Google environments scheduled for future release.
Research published in the fifth annual Cohesity Global Cyber Resilience Report highlights widespread enterprise vulnerability in this area. According to the findings, 56% of organizations feel unprepared to detect or contain unintended actions by AI agents, while 58% lack confidence in verifying the integrity of AI models and associated data following a security incident.
“Detection can tell you an AI agent went off course. It cannot undo the changes,” stated Vasu Murthy, Chief Product Officer at Cohesity. “Cohesity Agent Resilience provides a precise recovery path for both the agents and the resources they impact. Our vision for Autonomous Cyber Resilience is to use agentic workflows to reduce manual work from response and recovery.”
Complementing the new product release, Cohesity launched the Cohesity AI Resilience Academy, featuring an initial self-titled introductory course designed to educate organizations on mitigating operational risks as AI systems shift from advisory roles to active execution. The company also detailed broader updates to its Autonomous Cyber Resilience framework, integrating Cohesity RecoveryAgent and upcoming integrations with Cohesity Maestro to connect data protection workflows with customer AI tools such as Claude, ChatGPT, and Gemini.