My May 2026 best business reads roundup

/ AI is working. The value still isn’t 🙃

Your AI Budget Is Growing. Your Returns Aren't. Here's Why – Michael Heric, Purna Doddapaneni, and Antoine Débarre, Bain & Company

If you work in an enterprise context like I do, don’t skip this.

Finally some grounded numbers on why AI keeps underdelivering.

What I liked: the diagnosis is not “fix your data”. It’s organisational.

Hot paragraph:

“The technology worked. The value didn't arrive. And rather than pausing to understand why, 90% of those same companies are now increasing their budgets again… In reality, it is a circular bet with a structural leak.”

The peril of laziness lost – Bryan Cantrill

A sharp take on why fewer constraints push teams to stop optimising for simplicity: we get more output but worse systems.

This Is Why You’re Drowning in Busywork – Carl Benedikt Frey, The New York Times

AI makes it easier to do everything ourselves. It's satisfying, but it is pushing towards what I would call a productivity homeostasis: it overloads us with busywork without our noticing.

/ When AI changes how we build

“Vibe coding” is accelerating the erosion of design authority – Michael Buckley, Medium

Good piece on what happens when AI-generated prototypes are treated as answers, when they should still be questions.

This is not the usual “vibe coding is bad” take, I promise.

/ Early signals worth watching

The death of brand shitposting: why Gen Z now craves substance over sarcasm – Amanda Wallace, The Drum

"The old playbook was about capturing attention. The new one is about earning it". I don't know, it still feels a bit optimistic to me. But I want to believe in this shift.

Bonus read:

Everlane was never your friend – Andi Zeisler, Salon.com

SHEIN's acquisition of Everlane is the last nail in the coffin of the 2010s "ethical direct-to-consumer" era. It's also a useful case study of our relationships with brands.

(Photo of the month: Pope Leo XIV presented his first AI encyclical alongside Anthropic co-founder Chris Olah at the Vatican, on May 25th. A good reminder that the hardest questions around AI are no longer technical)

Originally shared as a LinkedIn post. Check it out here.

Next
Next

My March 2026 best business reads roundup