
When Will Enterprises Actually Use Agentic AI in Production?
Eye on AI4 July 2026Watch on YouTube
Description
Everyone's asking when AI agents will actually be running real business workflows at scale. Here's one of the most specific and honest answers we've heard. Alberto Pan, CTO of Denodo, makes a clear distinction: if you're asking when every large enterprise will have agentic AI in every relevant process - that will take a long time, because the remaining bottlenecks aren't technology anymore. They're organizational. They're legal. They're structural. But if you lower the bar to the most advanced 25% of companies using agentic AI in roughly 20% of their critical production workflows? His answer: by the end of 2027. And he thinks that might even be conservative, given how fast things are moving right now. Do you think enterprise AI will reach that threshold by 2027 - or does that feel too soon? Full conversation with Alberto Pan, Chief Technical Officer at Denodo, on Eye on A.I.
What you'll learn
- Technology is no longer the main barrier to agentic AI adoption, organizational and legal obstacles are more significant
- The most advanced 25% of large enterprises will likely deploy agentic AI in roughly 20% of their critical production workflows by end of 2027
- Agentic AI implementation in production happens unevenly, with early adopters among advanced companies deploying it first