42:I[185,["/_next/static/chunks/3xnt26d00_zk-.js","/_next/static/chunks/1s-yni1uf2lwd.js","/_next/static/chunks/42spz6paskvb6.js","/_next/static/chunks/06x30xgfe6iur.js","/_next/static/chunks/3b4qrzgkt_l12.js","/_next/static/chunks/15lgs8s22dk5l.js"],"SendEmailButton"] 44:I[63625,["/_next/static/chunks/3xnt26d00_zk-.js","/_next/static/chunks/1s-yni1uf2lwd.js","/_next/static/chunks/42spz6paskvb6.js","/_next/static/chunks/06x30xgfe6iur.js","/_next/static/chunks/3b4qrzgkt_l12.js","/_next/static/chunks/15lgs8s22dk5l.js"],"AddToLibraryButton"] 45:I[29721,["/_next/static/chunks/3xnt26d00_zk-.js","/_next/static/chunks/1s-yni1uf2lwd.js","/_next/static/chunks/42spz6paskvb6.js","/_next/static/chunks/06x30xgfe6iur.js","/_next/static/chunks/3b4qrzgkt_l12.js","/_next/static/chunks/15lgs8s22dk5l.js"],"ShareButton"] 46:I[58016,["/_next/static/chunks/3xnt26d00_zk-.js","/_next/static/chunks/1s-yni1uf2lwd.js","/_next/static/chunks/42spz6paskvb6.js","/_next/static/chunks/06x30xgfe6iur.js","/_next/static/chunks/3b4qrzgkt_l12.js","/_next/static/chunks/15lgs8s22dk5l.js"],"ProcessEpisodeButton"] 47:I[45269,["/_next/static/chunks/3xnt26d00_zk-.js","/_next/static/chunks/1s-yni1uf2lwd.js","/_next/static/chunks/42spz6paskvb6.js","/_next/static/chunks/06x30xgfe6iur.js","/_next/static/chunks/3b4qrzgkt_l12.js","/_next/static/chunks/15lgs8s22dk5l.js"],"SignupCTA"] 3a:["$","time",null,{"dateTime":"2026-08-06T10:05:01.000Z","children":"August 6, 2026"}] 3b:["$","span",null,{"aria-hidden":true,"children":"·"}] 3c:["$","span",null,{"children":"00:15:46"}] 43:T4c0,

Everyone wants AI in production. Very few are talking about the biggest thing holding it back. I had a great conversation with Matthew Carroll, CEO and Co-Founder of Immuta, at the Databricks Data + AI Summit, and one message came through loud and clear: AI doesn’t scale if people can’t securely access the data they need!!!!

We discussed why Immuta has evolved from being known as a data security company to focusing on data provisioning in the AI era.

Some of the topics we covered:

* Why many AI projects don’t fail because of the model, but because teams can’t get access to the right data at the right time

* What it really means for an enterprise to become agent-ready

* Why manual data access requests are becoming one of the biggest bottlenecks for AI adoption

* And the practical steps data leaders can take today to move from slow approval processes to policy-driven access

As more organizations move from AI pilots to production, conversations like these are becoming increasingly important.

The full interview is now live.

#data #ai #databricks #immuta #theravitshow