The race to preserve human judgment as AI rewrites financial decision-making
I was born in the late 1960s, which places me squarely in Generation X: the generation that grew up in the analogue world and helped build the digital one. We remember when financial infrastructure was visible — branches, paper instructions, clearing windows and people who knew why a control existed. We also helped create the internet economy, mobile banking, cloud computing, modern payments and the always-connected workplace.
That bridge role matters now because two historic transfers are colliding. The first is financial: UBS estimates that $83 trillion will move globally over the next 20 to 25 years. Cerulli Associates estimates that $124 trillion will change hands in the United States by 2048. The estimates use different methodologies and should not be combined, but their direction is unmistakable. The second transfer is less visible and potentially more consequential: judgment, institutional memory, relationship equity and decision-making authority are moving from experienced people into algorithmic infrastructure.
When I first wrote about the Great Wealth Transfer, I argued that it was never only about inheritance. It was about wealth, knowledge and the human judgment required to administer both. That remains true, but AI has accelerated the clock. For central banks, the question is no longer simply how individual institutions adopt new technology. It is how a financial system preserves trust, resilience and public purpose when intelligence becomes embedded in the infrastructure through which money moves.
Intelligence is Becoming Infrastructure
The cost of intelligence is collapsing. Stanford’s 2025 AI Index found that the cost of querying a model performing around GPT-3.5 level fell from about $20 per million tokens in late 2022 to roughly seven cents by October 2024 — a decline of more than 280-fold. McKinsey has estimated that generative AI could create $200 billion to $340 billion in annual value across banking alone.
That is an extraordinary opportunity. AI can improve fraud detection, supervision, risk analysis, customer service, compliance and operational efficiency. It can lower costs and widen access. But as intelligence becomes cheaper and more ubiquitous, context and accountability become more valuable, not less. A model can replicate a decision path; it cannot be assumed to possess the lived memory behind it.
Central banks understand this distinction instinctively. Monetary and financial stability depend on more than technically correct outputs. They depend on confidence, incentives, institutional behaviour and decisions made under stress. What happened the last time liquidity evaporated? Why was a payment control introduced? What vulnerability did a near-miss expose? When does a model’s precision conceal an incomplete view of the system? Those are not nostalgic questions. They are systemic-risk questions.
The Stewardship Test
A central bank is not simply another technology adopter. It is a steward of monetary and financial infrastructure: the arrangements that support settlement, liquidity, payments and public confidence. That stewardship role becomes more important as instant payments, tokenized assets, stablecoins and embedded finance compress the distance between decision and execution.
Speed brings genuine public benefit. Faster payments can improve cash-flow management, competition and inclusion. Tokenization may reduce frictions and enable new forms of exchange. Yet speed also shortens the time available to identify fraud, correct an error, manage liquidity or stop contagion. Settlement finality gives certainty to the system, but it also means that a mistaken or manipulated instruction may become irrevocable almost instantly.
This is why friction is not always the enemy. Some friction is …
Download your sample issue here!
No part of Central Bank Payments News may be reproduced, copied, republished, or distributed in any form or by any means, in whole or in part, without the express and prior written permission of the publisher, Currency Research Malta, Ltd.
