supervision
-
Fireside at AI-Native Banking and Fintech Pre-Conference
On September 29, I had a fireside chat with Madeline Fredin of Alloy Labs at a pre-conference event for industry and regulatory leaders attending SpringLab’s AI-Native Banking and Fintech Conference in Salt Lake City, Utah. We discussed AI risk governance, model risk management, evals, fintech partnerships, impacts on community banking, and regulatory training, amongst other Continue reading
-
Panel on “AI Adoption: Enablers and Blockers”
On September 24 in Frankfurt, I joined Dovile Grigiene (Revolut), Santa Purgaile (Bank of Latvia), and Alexander Thamm, for a panel discussion of AI adoption at the ECB’s annual Supervision Innovators Conference. We discussed the proof-of-concept ‘trap’, the role of culture and inertia, and the risks of moving too slowly. Continue reading
-
Publication of “Cybernetic and Superintelligent Supervision”
On July 22, 2025 the Forum on Financial Supervision, which is part of the London School of Economics’ Systemic Risk Centre, published my essay, “Cybernetic and Superintelligent Supervision“. The essay highlights that banking supervision’s ability to be effective is being challenged by long-term, irreversible trends. To adapt, supervision needs to cover more ground, better, and Continue reading
-
Workshop on “Cybernetic Supervision” for the ECB’s Technology and Innovation Division
On July 1, 2025, I led a two-hour virtual workshop on “Cybernetic Supervision” for the ECB’s Technology & Innovation Division (TIN). The workshop built off of my March 20 presentation to the ECB Supervisory Board, focusing on operationalizing the ideas presented. The workshop included live demos of publicly-available AI tools — from deep research to Continue reading
-
Presentation on “Cybernetic Supervision” to the ECB Supervisory Board
On March 20, 2025, at the invitation of Claudia Buch, Chair of the ECB Supervisory Board, I made a presentation on “Cybernetic Supervision” to the full Supervisory Board in Frankfurt. I framed supervision as a feedback mechanism (see graphs below), where the challenges of minimizing both “type 1 and type 2 errors” are heightened, especially Continue reading
Mike Hsu 