BMC Software has introduced AI tools for investigating mainframe problems and coordinating enterprise workflows, extending its established systems management products while retaining operational controls over automated activity.
The 20 October 2026 release covers BMC AMI Agentics, which supports database and mainframe administration, and new capabilities within Control-M, the company’s platform for managing application, data and infrastructure workflows.
BMC Software is developing specialised agents that use information about mainframe systems to investigate faults, explain probable causes and recommend corrective actions.
BMC AMI Agentics for Data focuses on IBM Db2 and IMS database operations. Its capabilities include advice on database reorganisation, analysis of SQL workloads and assistance with identifying deadlocks and database exceptions.
The company identified AI Reorg Advisor, SQL Workload Analyser for IBM Db2, Deadlock Advisor within BMC AMI Log Analyser and IMS Exception Advisor within BMC AMI Database Advisor among the components of the offering.
These functions address established database administration tasks. SQL analysis examines the behaviour of queries and their effect on system performance, while deadlock investigation concerns competing transactions that prevent one another from completing.
BMC AMI Agentics for Ops serves mainframe systems teams, with functions for investigating CPU usage, interpreting operational alerts and understanding automation written in REXX.
REXX scripts remain part of many mainframe environments, including processes developed and maintained over several years. BMC intends its agents to help engineers interpret that existing automation alongside current information about system activity.
The tools are designed around mainframe-specific knowledge rather than relying solely on a general-purpose AI assistant. They support investigation and recommendations, with decisions and operational actions governed by the permissions and processes used by the organisation.
BMC chief technology officer Ram Chakravarti said the company was concentrating on applying AI within controlled operational workflows, where administrators need visibility of the information supporting a recommendation and the ability to intervene.
The release also includes BMC AMI Service Discovery, which supplies verified z/OS configuration information to ServiceNow and other enterprise configuration management databases. The integration is intended to connect mainframe infrastructure records with the systems used to manage wider enterprise technology estates.
BMC AMI Assistant provides access to mainframe knowledge through enterprise collaboration tools, making information about specialist systems available within established operational communications.
Separate development functions remain in beta. BMC AMI DevX Unit Test can examine COBOL source code and copybooks in Visual Studio Code to generate synthetic test data and scenarios for developers.
A second beta capability links supported AI coding assistants, including Cursor, GitHub Copilot and Claude Code, with BMC AMI DevX Code Pipeline.
These integrations are intended to support code generation, promotion and deployment through existing development procedures. BMC says users remain subject to their mainframe permissions and that activities continue to be recorded through the established audit trail.
The beta designation applies to those developer capabilities. The company has not presented them as generally available production features across its customer base.
Control-M provides the other part of the October technology changes. The platform coordinates jobs and workflows across enterprise applications, databases, cloud systems and infrastructure, including processes in which one task must finish before another can begin.
BMC is extending its orchestration capabilities to support AI-driven activities alongside those established workflows. Natural-language monitoring and troubleshooting functions are intended to help operations staff examine dependencies, understand failures and determine appropriate recovery actions.
Data Assurance adds checks on information passing through workflows, allowing problems to be identified before affected data reaches downstream analytics or AI applications.
BMC’s announcement describes Data Assurance support within Control-M SaaS and self-hosted environments. Its accompanying technical commentary specifically identifies native Data Assurance within the SaaS release, so deployment details remain dependent on the product and version in use.
The company also named new integrations involving Amazon Bedrock AgentCore, Argo Workflows, SAP Build Process Automation, Workato, IBM MQ, Oracle JD Edwards EnterpriseOne, AWS Auto Scaling Groups and Google Cloud managed instance groups.
These connections extend Control-M’s ability to coordinate work across different systems. They do not establish that every integration is installed by existing customers or that AI processes can operate without their configured permissions.
BMC is using LangGraph technology within its agent development, while Control-M provides orchestration and operational monitoring for AI-related workflows.
The release brings together mainframe investigation, database analysis, developer assistance and workflow management. BMC has not disclosed adoption figures, commercial revenue associated with the changes or independently measured productivity improvements.
The new capabilities extend the functions available across BMC’s products, while deployment arrangements and the maturity of individual features differ. The company’s beta developer tools remain distinct from established operational services and the Control-M functions described in its October release.
You must be logged in to post a comment.