Every board deck this year has a line item for agentic AI. Far fewer come with a clear answer for what these systems are actually good at, and what they're not. In this episode, Russell Halper, Founder and Marketing Director at Insight Kitchen, sits down with our host David O'Keefe to untangle a technology stack that gets blurred together more often than it should: AI, large language models, optimization, and agents. Each solves a different kind of problem, and treating them as interchangeable is where investment decisions start to go wrong. The conversation covers why organizations are pouring money into agentic AI and where optimization actually fits inside that surge, the honest limits of what AI still handles poorly, particularly large scale numerical decisions and complex mathematical trade-offs, and the reality check executives need: the same investment rigor applied to any other major initiative, applied here too, with a clear look at where the returns are already strong, including in optimization. It also turns to people: how AI is reshaping consulting and expertise, why judgment, context, stakeholder management, and leadership still matter, and why keeping humans in the loop, to guide systems, review outcomes, and manage risk, becomes more essential, not less, as these systems get more capable. This is a conversation for leaders who want their AI investment grounded in what the technology can actually deliver, not just what it promises.