Five AI Questions We're Hearing from Financial Services Leaders

Summary

Financial services leaders are shifting from asking whether AI works to asking whether it can earn trust, amid governance and audit concerns. The post frames five operational questions around moving finance from reporting to action, using AI in financial crime without losing control, supporting bankers and investment teams, using real-time data, and managing expanding costs. Examples include governed GL-to-Risk reconciliation with cited evidence, human maker-checker oversight and an audit trail; AML agents that draft SARs while humans file; and multi-agent workflows for client and portfolio analysis. The post also cites Coinbase's sub-100ms P99 fraud-detection benchmark and warns that token costs manageable in pilots may grow in production, making model routing and spend visibility relevant. Databricks presents these topics through live demos, talks and meetings at Sibos 2026 in Miami, scheduled for September 28–October 1, including a session on governing AI in financial-crime operations.

Takeaways

1. 1

   Grant Thornton's 2026 Banking Insights AI Impact Survey found that half of banking executives said governance and compliance were limiting AI performance, while only 18% were confident they could pass an independent audit of AI controls.

2. 2

   The AML demonstration limits an agent to data the analyst is cleared to see, drafts a Suspicious Activity Report, leaves filing to a human, and logs each step against KYC/AML evidence and the model used.

3. 3

   The GL-to-Risk reconciliation demonstration identifies data breaks that can inflate risk-weighted assets, investigates them with cited evidence, and keeps a human maker-checker and examiner-followable audit trail in control.