I’ve noticed a recurring theme in my conversations with studio heads recently: their teams are evaluating AI production platforms and nobody is asking about the security architecture.
Sure, they’re asking about model quality; they’re asking about speed; they’re comparing UI features and watching demos with beautiful outputs.
Fair enough.
But savvy leaders are appalled at the lack of IP protection and security fundamentals in these products … and they probably should be.
I’ve spent my career at the intersection of security and production technology. I know all too well that a single leak of confidential IP could eviscerate trust and end companies. Security and reliability have never been optional for me; they’re the reason any of the other features matter.
What I’m seeing right now is an industry that is moving fast on capability and slow on the foundational question: when something goes wrong, who is responsible for what?
I wrote a longer piece on why I think this is the most important vendor qualification question studios aren’t asking yet.
Check it out and let me know your feedback!
Originally shared on LinkedIn, April 17, 2026.