Most companies are racing to adopt generative and agentic AI — but many are skipping the most important steps. In this episode of Today in Tech, host Keith Shaw is joined by Prem Natarajan, EVP, Chief Scientist, and Head of Enterprise AI at Capital One, for a candid discussion about what true AI readiness looks like in the enterprise. Learn why flashy demos aren’t enough — and what organizations must invest in to move beyond experimentation: Why data quality, cloud architecture, and talent are non-negotiables How Capital One built its own AI agent platform (and why most companies can’t yet) The difference between outputs vs. outcomes in agentic AI What “human in the loop” really means in 2025 Why trust must be earned — by the technology The overlooked role of governance, risk, and strategic patience AI success starts long before the first model is deployed.