Six billion dollars of experience
In 2012, a quant named Patrick Hagan built JP Morgan's portfolio risk model in Excel. He'd never built one before. It cost the bank $6.2B.
Hagan wasn’t an amateur. He co-invented a formula the whole industry still uses. But he had never built a value-at-risk model (which is a bit like asking a Michelin chef to fix the dishwasher). The model went live with a formula that added when it should have averaged.
Hagan didn’t lack skills or intelligence.
He had 30 years of experience.
Just not in risk management.
Benedict Evans recently said roughly "giving everyone a new way to make tools doesn't mean everyone will make tools." Most people just want to do their job. AI builders aren't selected for knowing the work best. They’re selected (or self selected) for being excited about the tech.
Put professional programmers and business users side by side on a no-code platform and the difference in correctness is statistically nothing. Building isn't hard. What's hard is knowing what edge cases won't show up in the pilot. Where this broke last time. Which shortcut looks fine until Q3. What a number looks like when it's wrong.
That knowledge sits with people who've done the work for years, and those folks rarely race to join a company's "AI Builders Cohort."
We don’t need more gates. (DO NOT SLOW THE BUILDERS DOWN.) We need more experienced hands inside the build. Pull the veterans who'll have to live with the thing into making it. Not reviewing it. Not poking at a prototype. Making it.
Two things happen at once: The expertise gets built in while the shape is still soft, instead of bolted on later. And the people who would otherwise be targets of an adoption campaign become the reason the thing gets adopted at all. (Behavioral economists call that the IKEA effect: people love the things they create).
Every model that blew up had someone who could have said "that number looks wrong."
They just weren't in the room while it was being built.
—
Sources:
JPMorgan Chase Whale Trades: A Case History of Derivatives Risks and Abuses, U.S. Senate Permanent Subcommittee on Investigations, 2013
The IKEA Effect: When Labor Leads to Love, Norton et al, 2011
Another Podcast, Jul 24, 2026
"Low-code vs. the developer: An empirical study on the developer experience and efficiency of a no-code platform," Guthardt et al, 2024