BMC Software has announced BMC AMI Agentics, a set of AI agents for mainframe database and operations work, together with new Control-M capabilities for orchestrating AI, data and application workflows. The company says the aim is to put AI to work in critical operations while people stay in control.
BMC AMI Agentics for Data helps teams investigate Db2 and IMS issues, explain findings and guide performance decisions, with tools such as AI Reorg Advisor, SQL Workload Analyser for IBM Db2, Deadlock Advisor and IMS Exception Advisor. BMC AMI Agentics for Ops helps operations teams investigate CPU spikes, make sense of legacy REXX automation and identify probable causes within alerts.
Two further mainframe features are in beta. BMC AMI DevX Unit Test parses COBOL source and copybooks inside VS Code to create synthetic test data and scenario paths. Separately, supported AI coding assistants including Cursor, GitHub Copilot and Claude Code can connect to BMC AMI DevX Code Pipeline, so developers can generate, promote and deploy mainframe code from their chosen assistant, under their own mainframe permissions and with actions recorded in the existing audit trail.
On the Control-M side, Data Assurance validates data as it moves through business workflows, in both Control-M SaaS and self-hosted versions, to catch quality problems before they reach analytics, reporting and AI projects. AI monitoring adds natural-language troubleshooting of job and workflow dependencies. New integrations include Amazon Bedrock AgentCore, Argo Workflows, SAP Build Process Automation, Workato, IBM MQ, Oracle JD Edwards EnterpriseOne, AWS Auto Scaling Groups and GCP Managed Instance Groups. BMC also uses LangGraph inside BMC AMI Agentics, and Control-M integrates with LangGraph.
Ram Chakravarti, chief technology officer at BMC, said: “AI must earn the right to run your business.” He said AI earns trust in critical operations by being open by design, governed by default and human-controlled.
Brent Ellis, vice president, AI and Workloads at IDC, said: “Governance, not model capability, is what separates AI pilots from reliable execution. Closing that gap requires context, orchestration, and control across complex enterprise estates.”
To stay across the latest in cloud, AI and enterprise tech analysis from Compare the Cloud, subscribe to our weekly newsletter at https://www.comparethecloud.net/newsletter
