Clear Perspective: AI Boundaries

AI doesn't stop at the edge of a test environment — and that's the problem. 

In this episode, Harry Lu, VP of AI Solutions, shares what recent AI safety incidents reveal: misaligned goals, excessive access, and systems relentless enough to blur the line between testing and production. Harry makes the case for treating every AI deployment — tested or live — with the same guardrails, oversight, and readiness to respond. 

For healthcare organizations adopting AI fast, the message is simple: set the boundaries now, before the frontier models become tomorrow's norm.