The AI in Business Podcast Why Financial AI Can't Scale Without Unified Governance with James Dean of Google and Mark Crean of Securiti
7 snips
Mar 18, 2026 James Dean, AI Specialist at Google Cloud who helps banks apply cloud and AI with strong governance, and Mark Crean, Regional VP at Securiti focused on data security and privacy, discuss why fragmented sensitive data stalls finance AI. They cover operational data bottlenecks, aligning security and business teams, automating classification and controls, and building centralized governance to scale AI safely.
AI Snips
Chapters
Transcript
Episode notes
Data Location Is The Fundamental AI Bottleneck
- AI scale depends less on models or chips and more on knowing where sensitive data lives and who can access it.
- Mark Crean emphasizes questions like data location, sensitivity, access, last access time, and compliance frameworks as foundational.
Siloed Data Keeps Pilots From Production
- Data silos are keeping about half of banks' AI pilots trapped in labs rather than production.
- James Dean cites reports showing roughly 50% of banks still operate with siloed data, blocking scale.
Classify Data Before Training Models
- Automate data classification before training or cloud migration and embed strict access controls into model operations.
- James Dean gives the example of NLP scanning unstructured docs to tag hidden KYC/PII so models never train on restricted info.