BMC: Компании переходят от экспериментов к операционному использованию ИИ на мэйнфреймах

BMC Survey Shows Shift to Operational AI on Mainframes In its 2026 Mainframe Survey released this week, BMC revealed how companies are using AI with mainframes. The data indicates a clear change in how businesses use this technology, with the focus moving from testing AI to using it in daily work. The survey gathered answers from more than 1,300 professionals and decision-makers around the world. The results show that mainframe technology remains central to business. About 94% of respondents said they have long-term confidence in the platform and said they plan to continue investing in it. For 45% of those surveyed, implementing AI is a top priority. The Move to Trusted AI For years, AI on the mainframe was often experimental. That phase is ending. Companies now view AI as an operational tool. However, the data shows that this adoption is pragmatic. Organizations are cautious. They want AI to provide insight and advice, but they are not ready to give it full control. This cautious realism is evident in the survey numbers. For example, 40% of respondents are willing to have AI suggest actions for code management, yet only 23% are comfortable letting AI complete those actions on its own. A similar trend appears in database reorganization. About 43% of respondents want AI to recommend actions, an increase from 37% the previous year. Despite this interest in recommendations, only 21% of respondents are comfortable letting AI finish the task without human help. Goals and Hurdles Companies are prioritizing AI initiatives that offer clear benefits – improving productivity and simplifying their operations. They also want to use AI to preserve knowledge and speed up the modernization of their systems. Common tasks for AI include performance tuning, problem detection, managing IMS queues, and generating documentation. Despite these goals, teams face obstacles. The survey highlights several concerns that slow down implementation: - Implementation costs: 41% of respondents cite high costs. - Security and privacy: 39% identify these as major risks. - Data integration: 37% note that moving and using data is difficult. - Regulatory and compliance rules: 22% find these requirements a barrier. Future Investments and Security Companies are looking ahead to agentic management. This involves using AI agents to manage parts of the mainframe. The survey shows that 36% of organizations plan to build their own agents. Another 32% plan to buy agents from third-party vendors. Security management is also a specific area of focus. New rules require reducing the standard digital TLS certificate life cycle to 88% by 2029. This change will create more work for IT teams, and if they don’t have a clear plan for that, these teams could face service disruptions. Currently, certificate management is split across the industry. About 43% of shops use automated solutions they built in-house. Another 31% use commercial, product-based automated solutions. However, 25% still rely on manual effort. This gap represents a risk as new compliance requirements take effect. The Human Role John McKenny is the senior vice president and general manager of Intelligent Z Optimization and Transformation at BMC. He said the industry is moving past the stage of simple experimentation. He notes that companies have stopped asking, “How can we use AI?” and are now asking, “Where can we trust it?” McKenny believes that the path to AI autonomy on the mainframe depends on trust. He said that businesses need to know where AI can be governed and where it delivers real value. He emphasized that the option for humans to stay in the loop will remain necessary. Humans will continue to oversee and implement recommendations from AI tools.