Data Access and Other Bottlenecks in Enterprise AI Adoption: DJ Patil Live with Tim O’Reilly

DJ Patil co-coined the term "data scientist," served as America's first chief data scientist under President Obama, was chief scientist at LinkedIn, and has spent the past decade on the founding team at Devoted Health, where he's built the kind of data infrastructure that most organizations are still struggling to create. DJ’s been on a listening tour. Wherever he travels, he finds a local university, holds office hours, and asks whoever shows up—students, faculty, hospital administrators, executives—what they're actually experiencing with AI. What he's hearing about the anger, angst, and the “impasse of dialogue” between AI boosters and skeptics is helping him refocus his thoughts about AI adoption. He joined Tim for a wide-ranging conversation about what's working right now, what's broken, and where the real bottlenecks are.DJ and Tim started with students, who are feeling that the social contract of “go to college and start a career” is broken, and DJ's plan to launch a makerspace-style program for those who didn’t land internships this summer to learn and demonstrate their skills. Then they went deep on why the organization is the bottleneck to transformation, using the healthcare industry as an example of the entrenched challenges and what’s possible when you get the infrastructure right. DJ walked through how Devoted Health built its data foundation before LLMs existed, why that tidy house is now a compounding advantage, and the change we can make when transforming our healthcare system is, as he put it, "like walking, chewing gum while balancing bowling balls on your head and on a unicycle.""We're [both] this giant human LLM," DJ told Tim, "summarizing and distilling what we're hearing from a lot of people." What they're hearing is that the chief constraint is whether our institutions can build the organizational and economic infrastructure to actually deploy what we've built.