Rocket Software unveils EVA 2.0, its next-generation Enterprise Virtual Assistant platform designed to leverage AI for deeper mainframe operational insight and controlled automation while preserving strict security governance.

  • EVA 2.0 introduces PlanGuard to mediate AI agent actions with contextual policy checks.
  • Platform preserves existing mainframe security controls while enabling AI-driven investigations.
  • Pilot deployments demonstrate reduced problem investigation times in financial and retail sectors.

What happened

Rocket Software announced the upcoming release of EVA 2.0, an enhanced version of its Enterprise Virtual Assistant platform. This update adds PlanGuard, a new security framework designed to govern AI agent actions on mainframe systems. PlanGuard dynamically evaluates each proposed AI operation based on user identity, organizational rules, and environmental context before allowing or denying execution.

The platform aims to advance automation of mainframe operations by enabling multistep AI-led investigations initiated through natural-language requests. EVA collects and correlates operational data from multiple sources to provide clear explanations and insights. PlanGuard further restricts AI capabilities by creating temporary, task-specific execution identities and maintaining a comprehensive, tamper-evident audit trail for all agent activities.

Why it matters

Mainframe systems remain critical for many enterprises, especially in regulated industries where security and auditability are paramount. EVA 2.0’s PlanGuard addresses a key challenge for AI adoption in these environments: how to balance automation benefits with strict governance and compliance requirements. By integrating with popular mainframe security tools like RACF, ACF2, and Top Secret, it preserves existing controls while introducing fine-grained, contextual authorization.

This approach enables organizations to harness generative AI to reduce operational complexity and investigation times without risking unrestricted AI access to sensitive resources. The detailed audit trail supports transparency and accountability, empowering security teams and auditors to verify the rationale and compliance of AI-driven actions.

What to watch next

Rocket’s pilot programs across financial services, government, insurance, retail, and telecommunications will reveal how effectively EVA 2.0 and PlanGuard improve operational efficiency and security. Key metrics to watch include reductions in problem resolution times and user adoption levels by mainframe operations teams.

Future developments may expand EVA’s automation capabilities beyond analysis to more autonomous system changes while maintaining strong governance. The broader acceptance of agentic AI in mainframe management will depend on how well Rocket balances innovation with risk management through tools like PlanGuard.

Source assisted: This briefing began from a discovered source item from SiliconANGLE. Open the original source.
How SignalDesk reports: feeds and outside sources are used for discovery. Public briefings are edited to add context, buyer relevance and attribution before they are published. Read the standards

Related briefings