Relevance Reboot: The Half-Life of Expertise

In my last essay, I wrote about AI’s Galileo moment – how intelligence turned out to be less scarce than we pretended, and relevance now lives in judgment rather than recall. But that philosophical shift raised a practical question for me: why does maintaining relevance feel so much harder now than it did just a couple of years ago? It’s not just that the bar moved. It’s that the clock itself changed.

---

Moore’s Law didn’t just measure chips. It measured careers. Performance doubled. Skills aged predictably. You could learn something deeply, apply it for years, and expect that it would remain relevant long enough to matter.

That clock didn’t just fade. It broke.

This recently hit home for me while trying to understand autonomous coding agents well enough to think through their policy and security implications – what they are, how they work, and where the real risks live. I started by following the then trendy Clawdbot, reading about it, and trying to make sense of what it meant for companies. Before I’d even finished wrapping my head around the reporting, it was renamed Moltbot. Then, in less than 72 hours, it was renamed again to OpenClaw, with a public skills marketplace and hundreds of thousands of downloads in its first week. And that was just the product evolution. Every CISO conversation I’ve had in the past two weeks surfaced a different security model, a different deployment pattern, and a different set of concerns. Each iteration wasn’t wrong, and what I’d learned wasn’t obsolete. It was simply incomplete before I’d even finished learning it. And I’m not the only one.

AI hasn’t simply accelerated technology. It’s changed the unit of time itself. The clock didn’t just speed up. It started ticking differently. What once unfolded over years now unfolds over months, sometimes weeks, and occasionally days. Something that felt revolutionary last quarter can feel outdated today. Not because it was wrong, but because the terrain shifted.

That’s the new tempo.

Moore’s Law told us how fast things got better. AI is teaching us something far more uncomfortable: how fast relevance decays. The half-life principle fits. It’s not failure or disappearance, just slow erosion. What you knew six months ago may still be true – it just carries half the influence it once did. Where expertise used to compound, it now depreciates. Knowledge is no longer something you accumulate and bank. It’s something you maintain, like muscle, reputation, or trust.

Many experienced tech professionals are encountering this. The gap between “I just learned this” and “this has already changed” keeps narrowing. Many of us built our careers using maps. We learned the terrain, memorized the routes, calculated the time to travel, and knew where the hazards were. Updates didn’t happen very often because last year’s map was still usable. You could rely on it.

Today, we’re in a GPS world. Real-time updates, constant recalculation, and rerouting on the fly. There’s no loyalty to the route you took yesterday if conditions have changed today. The danger isn’t that paper maps were wrong. It’s insisting on using them and then wondering why the world feels disorienting.

This shift hits hardest for people who earned their reputations over decades. Experience still matters. Judgment still matters. So does scar tissue from failure. But legacy alone is no longer a strategy when time itself has sped up.

AI makes it easier to look smart and harder to be right. What matters now is knowing what questions to ask, what breaks at scale, what harms people, and what shouldn’t be automated simply because it can be. That kind of relevance doesn’t last on its own. You have to earn it again – in the present.

Relevance Reboot has never been about chasing every new technology or tool. It’s about thinking differently about time – shorter learning loops, lighter attachment to specific tools, and a willingness, especially later in a career, to feel like a beginner again. Not just once, but regularly. After decades in the tech arena, I find myself relearning things long in my rearview mirror, not because I got them wrong the first time, but because the context keeps shifting. The advantage isn’t knowing the answers. It’s recognizing the patterns when the details change and adjusting before others realize they need to. That’s the paradox I keep running into. Experience still matters, but only if you treat it as a compass, not a conclusion. The people struggling most right now aren’t the ones who lack experience. They’re the ones who believe they’ve already finished learning.

Time isn’t slowing down. But relevance is still available to those willing to keep moving while the clock is being rewritten. The route you took to get here will not be the route that keeps you here. That’s not a failure of your past work. It’s the condition of the present.

The question isn’t whether you built something valuable. It’s whether you’re willing to rebuild it, even when you’re tired of rebuilding.

---

This essay reflects my personal views and experiences and do not represent the views of my employer or any organization I am affiliated with.