When AI steps into the boardroom, it’s rarely a quiet arrival. Expectations are set high, with promises of sharper productivity, streamlined workflows, and measurable efficiency gains. But as Francesco Brenna, Global Leader of AI Integration Services at IBM Consulting, points out in this episode of Consulting the Future, the true opportunity lies in something far deeper than speed. It’s about reimagining the way entire businesses function.

Recorded during a sweltering summer in New York, our conversation breaks down what “agentic AI” really means for leaders under pressure to make AI more than a buzzword. Francesco draws a clear line between the passive assistants many companies have experimented with and the next generation of intelligent agents that not only advise but act. This shift, he argues, demands more than dropping AI into an existing system. It calls for rebuilding processes from the ground up with business outcomes as the starting point.

We dig into why data readiness remains the number one barrier to success, despite years of investment in platforms and governance. Francesco introduces the concept of “data products” to ensure AI agents operate with the right context and memory. He outlines IBM’s three-layer approach to agentic applications: user experience, orchestration, and data. He also explains why standards like Model Control Protocol (MCP) may be the key to integrating AI with legacy systems at scale without sacrificing security or trust.

Francesco shares real-world results from IBM’s work in customer service, insurance, and pharma, where agentic AI has dramatically improved containment rates, reduced months of work to weeks, and enabled smarter decision-making for knowledge workers. He is candid about the human side of adoption, detailing how IBM uses hackathons, hands-on experimentation, and human-centered design to build confidence and capability across the workforce.

For enterprise leaders grappling with how to move AI out of the pilot phase and into meaningful, measurable impact, this conversation offers a grounded, practical roadmap built on doing the right AI, and doing it right.