Six months ago the consensus take among engineering leaders was that nobody was going to just stop writing code themselves. Real engineers wouldn't accept generated code. The craft of typing was load-bearing for understanding. I held some version of this.
Most of us have stopped. "All steering, no handwriting" is how DHH - creator of Ruby on Rails - put it today. My own org ships 90%+ AI-written code. Your team is reviewing PRs they didn't write.
But this post isn't really about AI. It's about how groups update when reality moves faster than identity can - or how they don't. AI is just the case study where it's happening fast enough to watch in real time.
Build-up-to-tear-down is the oldest pattern in media: praise someone, then dismantle them. "We're so back" / "it's so over" is the meme version compressed to weeks. Smartphones took a decade to complete a full cycle and you couldn't see the mechanism. AI takes weeks and the release cadence is fast enough that you can watch a full cycle complete in real time.
What looks like a random walk between excitement and scepticism isn't random. The engine underneath is status, arranged like an onion. The higher-status, more-online edge holds a take first, it diffuses to the next layer, and then to the broader audience. Your stance is set by whoever's one layer out: when the layer below you is building it up, you tear it down and once "build up" has rolled two layers out and the layer below you starts tearing it down, you build it up again. The cycle isn't noise, it's gradient descent. The loss function isn't "be right", it's "don't sound like the layer below". The mechanism is distaste - each layer finds the layer just below it faintly embarrassing and the discourse is the structure that keeps distance. Karpathy's snapshot four months ago - double-digit engineer adoption against "low single digit" general awareness - is the onion in cross-section.
But while sentiment cycles at the surface, the ground keeps rising - the capability underneath isn't part of the cycle, it just accumulates. Each "we're back" lands further along than the last "we're back" did - same meme, higher floor. That's the ratchet. The discourse feels cyclical but looking back nothing has actually come back - the position just gets redrawn around a new substrate.
So what happens to a specific position when the ground rises out from under it? Mostly, people don't update. They migrate. The position climbs a ladder of unfalsifiability: from "can't" (easy to disprove) to "isn't really what we meant" (definitional) to "is bad" (pure value claim). It's a motte-and-bailey that keeps losing territory - the edge needs somewhere new to stand and "is bad" is a rung no screenshot can knock you off.
You can see this happening right now with the engineering cope. The same people who said engineers wouldn't stop typing have moved sideways: "typing code was never what engineers were paid for anyway." Same scepticism, repointed somewhere unfalsifiable. Others who held the same view never said it loud enough to need a retraction - they just quietly stopped saying it. Both moves work and nobody has to admit anything.
The same move has been running with AGI for a decade. Every benchmark that defined AGI in 2018 has been passed. None of them are called AGI now. The capability accumulates and the term migrates to wherever the capability isn't. That isn't bad faith. It's the natural defence of a sceptical position against the ratchet - it needs somewhere to live and it always finds a rung.
What does this give you? A heuristic. If a take feels safe to hold this quarter, it's probably already behind the ratchet. Where the discourse feels comfortable is exactly where it's wrong, because comfort is what the ladder evolved to provide.
Of course, writing this is itself the next rung. "AI is hype" has diffused, "AI is back" has diffused and "actually the discourse is all signalling" is the next edge position to hold - and I'm holding it now. The capability is real, the discourse is theatre - and I'm not above it myself. Either way, the ratchet is real.