Enterprise IT leaders are on the frontlines of AI transformation. As businesses invest heavily in technology, they’re responsible for its implementation. They’re tasked with delivering innovation at speed and generating cost savings, productivity gains and business growth.
But pressure is mounting, and problems are growing. In July, a report by MIT, titled The Gen AI Divide: State of AI in Business 2025, revealed that despite $30–40bn of enterprise investment in generative AI, 95% of organisations are seeing zero return.
The problem isn’t the technology or a lack of IT talent. It seems to be the approach adopted by businesses. The report found that 80% have explored or piloted GenAI tools such as ChatGPT and Copilot, and nearly 40% have reported deployment. But these tools were designed to enhance individual productivity rather than profit and loss performance.
The result is that generic AI tools are being used as add-ons to projects and processes and aren’t driving the meaningful strategic and structural changes needed to deliver returns. This is bad news for enterprise IT leaders who are increasingly being tasked with scaling smarter – enhancing what employees can accomplish without needing to increase or decrease headcount. To do that, leaders need to reset enterprise IT economics by using AI to create value, not cost.
Enterprise-grade agentic AI
AI agents could provide the solution. Agents are a software system that uses AI to autonomously perform tasks, reason, plan and learn to achieve goals on behalf of a user.
Ali Siddiqui is the President of BMC Helix, a firm that provides ready-to-use AI agents for core IT functions and enables businesses to create custom agents for their unique needs.
He says agents can help enterprise IT leaders to deliver laser-focused AI projects that solve specific business problems, multiply productivity and yield ROI.
“The massive shift needed, especially in IT, is really about delivering measurable business impact,” he says. “These supercharged AI agents, especially within enterprise software, are allowing CIOs to automate full business functions. Agents are capable of doing autonomous work with little or zero human intervention. This reduces the load stress on human employees and allows them to focus more on innovation and strategic work.”
The need to shift from generic AI tools like ChatGPT to enterprise-grade AI agents is clear. But the MIT report revealed that most custom or vendor-sold systems are being rejected. While 60% of organisations evaluated the potential of these tools, only 20% reached pilot stage and just 5% reached production. Brittle workflows, a lack of contextual learning and misalignment with existing daily operations were cited as major problems.
Building and scaling AI agents
But those that do partner with external providers are enjoying twice the success rate of internal builds. Siddiqui says the starting point for IT leaders is to identify profitable use cases and use their platform to build agents to solve them. “Internally, IT leaders need to win executive sponsorship, and so they must choose the right starting point,” says Siddiqui. “This means identifying a single, clear business function and use case that could be automated and deliver ROI. From there, you can scale your use cases – and if you’ve chosen a provider that specialises in that function, you’ll gain access to a fleet of out-of-the-box agents that are ready to go.”
IT teams are often overwhelmed with time-consuming, routine tasks that could easily be automated using BMC Helix’s GPT-powered fleet of out-of-the-box AI agents. Autonomous incident response agents can automatically detect, triage and resolve common IT issues, such as server downtime and configuration drift. Service teams can also improve their productivity by using agents to resolve frequent queries such as password resets and access requests.
Security is another area that IT teams can address. Agents can be built to identify costly, potential risks and boost resilience. By identifying vulnerabilities, recommending resolutions and simplifying root cause analysis, agents can reduce the number of change-related incidents. The completion of complex tasks, such as the analysis of organisational knowledge, is also enhanced by using agents to remove silos and extract insights from multiple departments to improve human decision making.
Human-centric strategies
But if fleets of AI agents are capable of carrying out much of the work of humans, what does this mean for the future career prospects of IT teams? It’s a question that Siddiqui faces on a regular basis. Here, education and communication are critical to ensure IT leaders derisk change and get buy-in. Not just from their CFO, but also from the workforce faced with a new future in which their co-workers won’t only be human.
“Leaders need to work with stakeholders to be clear about the use cases and how agents will enhance their working lives,” says Siddiqui. “First, agents will free them up to focus on projects that really show how they’re impacting the bottom line. Secondly, one of the biggest opportunities is reducing non-headcount spend on things like LLM and capacity costs. Agents can aid with discovery and enable IT teams to show clear KPIs around cost savings, as well as revenue generation, during IT transformation projects.”
Hundreds of businesses are already embracing the power of BMC Helix’s AI agents. One of the company’s biggest clients processes nearly three million HR cases per year. But autonomous AI agents enabled the customer’s team to access immediate summaries of cases (a small but repetitive task). Likewise, this customer’s support desk employees can now get quick answers to questions regarding their cases using an agent called Helix GPT Service Collaborator. Agentic support of the support experts multiplies their productivity by delivering actionable insights and offering advice for faster incident resolution. Overall, Helix GPT saved the customer $2.9m per year.
Agents are also helping businesses to predict the future. Network operations centres (NOC) are command centres where IT teams monitor, manage and maintain networks, servers, applications and other critical IT infrastructure. Much of their time is spent dealing with incident management. But agents can now predict most issues long before they impact customer experience, helping IT leaders to improve customer satisfaction and retention. In doing so, they can scale performance and deliver the measurable ROI of AI.
To find out more about scaling with Agentic AI, please visit BMC Software
Enterprise IT leaders are on the frontlines of AI transformation. As businesses invest heavily in technology, they’re responsible for its implementation. They’re tasked with delivering innovation at speed and generating cost savings, productivity gains and business growth.
But pressure is mounting, and problems are growing. In July, a report by MIT, titled The Gen AI Divide: State of AI in Business 2025, revealed that despite $30–40bn of enterprise investment in generative AI, 95% of organisations are seeing zero return.
The problem isn’t the technology or a lack of IT talent. It seems to be the approach adopted by businesses. The report found that 80% have explored or piloted GenAI tools such as ChatGPT and Copilot, and nearly 40% have reported deployment. But these tools were designed to enhance individual productivity rather than profit and loss performance.
