The Punk PM #54

Hey there, punk!

The scaremongering is relentless. AI is taking your job. product management is dead. Anyone who hasn't pivoted to being an "AI PM" by last Thursday is already finished.

Lenny's State of the Product Job Market from March has a different take: product openings at their highest in more than three years. Not dead. Not even close.

What is dead, though, is a certain type of product management.

Let's get into it.

Quote of the Week 🙊

Insight 🦉

Nobody knows which jobs AI is going to kill. And honestly? Anyone who tells you otherwise is either selling something or hasn't thought hard enough about it.

We're so early in this that we can't even agree on what needs protecting, let alone how to protect it. The debate around AI regulation is full of confident voices drawing lines in the sand, but the sand keeps shifting. What we actually have is a lot of people with vested interests trying to freeze-frame a moment in a film that's still being shot.

This isn't new. Technology has always done this. It disrupts, displaces, creates, destroys. And the full picture only becomes clear in retrospect.

I left school in 1997. The job I do now barely existed then. Product management happened, sure, but it didn't become a career path people actually chose until the App Store and SaaS changed the economics of software. What I wanted to do was be a music journalist. And I did, for a while, freelancing for magazines that felt permanent and important. The internet had other ideas, though. It didn't just disrupt music journalism; it hollowed it out, devalued the craft, and scattered the audience across a million free alternatives. The role didn't disappear entirely, but the industry it sat inside basically collapsed.

The lamplighter is a better example. From the 16th to 19th century, there were people whose entire job was to walk the streets every evening, lighting gas lamps by hand. Essential work. And then electricity arrived and made the whole thing redundant almost overnight.

Here's the thing: what if we'd tried to protect that role? Imagine the legislation. Imagine a world where, right now, someone still has to physically switch on every electric streetlight in Newcastle every evening. Hundreds of people. Billions of pounds. All to preserve a job that electricity made pointless, and to slow down a technology that eventually powered the entire modern economy, including whatever it is you do for a living.

Product management has its own lamplighters. Writing tickets. Running refinement ceremonies. Maintaining the backlog. Producing weekly status updates nobody reads. A huge chunk of what many PMs spend their time on is coordination and documentation; the connective tissue of delivery. AI is already pretty good at it, though. And it's getting better fast. If your value proposition as a PM is "I write good user stories and I keep Jira tidy," that's a problem—not because you're bad at your job, but because the job itself is shifting underneath you.

Without electricity, none of our jobs exist. Not mine. Not yours. The lamplighter's displacement wasn't a tragedy to be prevented. It was the price of admission to everything that followed.

AI will do the same thing. The jobs at risk right now are real, and the disruption is real, and we should absolutely be thinking about how to support people through that transition. But there's a difference between protecting people and protecting roles, and we keep confusing the two.

The goal of regulation shouldn't be to preserve every job that currently exists. It should be to make sure the humans displaced by this shift land somewhere better, not somewhere worse. Retraining. Safety nets. Time to adapt. That's what good policy is for.

For product managers, the implication is more immediate. The parts of this job that are about synthesis, judgement, and commercial instinct—figuring out what to build and why, navigating stakeholder politics, making the call when the data doesn't give you a clean answer—those are harder to automate. The parts that are about process, ceremony, and documentation? Less so.

If you are the AI revolution's equivalent of a lamplighter—if your value is in execution that can now be automated—then no amount of regulation will change the underlying reality.

Don't wait to find out. You probably already know.

Action 🚀

Next week, list the three things you spend the most time on as a PM. Be honest, not aspirational.

Then ask: could AI do any of these reasonably well?

If the answer is yes, that's not a reason to panic. It's a reason to work out what you should be doing with the time instead.

Inspiration 💡

The Majority AI View – Anil Dash on the opinion you almost never hear: most engineers and PMs inside tech think AI is a useful technology being grotesquely overhyped, and they're afraid to say so. The loudest voices aren't the representative ones. The lamplighter narrative itself may be being overplayed by people with a financial interest in panic. Read more

Another Viral AI Doomer Article, the Fundamental Error – Ben Thompson on why most arguments about AI destroying jobs make the same structural mistake: they treat AI as a monolithic force rather than a specific technology with specific applications. The DoorDash case study is particularly sharp—AI winning in one context doesn't mean it wins everywhere. Precision matters more than prophecy. Read more

The Fall of the Nerds – Noah Smith on how Silicon Valley's cultural authority—the idea that tech people are benevolent problem-solvers making the world better—has collapsed under the weight of its own contradictions. The lamplighter analogy cuts both ways: it's not just individual roles being made redundant. The whole tech priesthood narrative is being switched off. Read more

Signing Off ✍️

If this resonates with you, hit reply and let me know. And if you think a friend or colleague would enjoy The Punk PM, feel free to share it with them!

Play it your way,

Toby