THIS WEEK’S SIGNAL

AI vendors are no longer selling into banking — they are embedding inside of it. FIS and Anthropic's joint announcement of agentic AI for financial crimes, with BMO and Amalgamated …

Read more

THIS WEEK’S SIGNAL

Customers Bank's OpenAI partnership is not a pilot. It's a production deployment to re-engineer commercial banking operations by end of 2026, targeting sub-20-minute account opening for complex commercial clients. Meanwhile, a new Mythos report reveals 69% of all financial services respondents rely on OpenAI (76% among banks), creating what regulators are now calling "notable critical third-party risk." The concentration risk is no longer theoretical.

DEEP DIVE

Customers Bank CEO Jay Sidhu deployed an AI clone of himself on the bank’s Q1 earnings call this week, then announced a direct partnership with OpenAI to build what the bank describes as “one of the first AI-native regional banks” by the end of 2026. This is not a sandbox. The bank is targeting three operational areas immediately: lending (AI handles document collection, credit memos, legal docs, post-closing portfolio monitoring), deposits (AI streamlines digital onboarding and compliance workflows), and commercial account opening (collapsing multi-day processes to under 20 minutes via conversational AI and automated document gathering). The stated objective: shift bankers from processing work to client-facing work within eight months.

Read more

THIS WEEK’S SIGNAL

Federal AI policy is fracturing in real time. Banks now face contradictory mandates: Treasury, Fed, OCC, and FDIC push AI adoption for fraud prevention and compliance, while defense…

Read more

THIS WEEK’S SIGNAL

Anthropic's Mythos AI model, launched April 7 under Project Glasswing, triggered simultaneous regulatory review across four continents and a partnership scramble among Tier 1 banks. Meanwhile, Bank of America shed 1,000 jobs in Q1 while increasing technology spend 10%. The gap between AI rhetoric and AI consequence just closed.

DEEP DIVE

Anthropic’s Mythos becomes the first AI model to force coordinated global regulatory response before widespread deployment. Asian regulators (ASIC, MAS, HKMA) issued monitoring statements on April 20 and 21, citing cybersecurity and systemic risk concerns tied to Mythos’s autonomous coding and decision-making capabilities. European regulators are in “close contact” with banks exploring access, according to a senior banker quoted by Reuters. Piraeus Bank (Greece) and Accenture launched a dedicated AI Hub powered by Claude on April 22, explicitly framed as a move toward agentic AI in core workflows.

Read more