The corporate world has a shiny new metric, and it doesn’t care about your tenure or your degree. From Accenture to JP Morgan, the biggest names in business are wiring AI proficiency directly into the promotion pipeline. Hit the leaderboard, get the bonus. Miss the training, clear your desk. It sounds like a meritocracy for the machine age, but for workers like Duncan Trevithick, it feels disturbingly like being handed the shovel to dig your own grave.
The Leaderboard Logic
Walk into a modern corporate floor — or log into the Slack channels of Disney, Meta, or KPMG — and you will find a new kind of scoreboard. These “AI leaderboards” track everything: prompt volume, tool adoption, completion of mandatory training modules. At Coinbase, Brian Armstrong didn’t just suggest his engineers learn the ropes; he fired the ones who didn’t. Accenture’s Julie Sweet put it bluntly on the Rapid Response podcast: AI is now “how we do work.” If you want to climb the ladder, you had better start climbing the rankings.
The pressure is palpable. With UK job vacancies hitting a five-year low, the leverage has swung decisively to the employer. A July poll by recruiter Gi Group found 75% of 1,881 British jobseekers wouldn’t be deterred by AI proficiency appearing in performance reviews. The message is clear: the wave is coming. You either grab a surfboard or you drown.
The Productivity Paradox
Here is where the gadget-loving optimist hits a hard wall of reality. Trevithick, a 34-year-old marketing lead at an AI training data firm in Spain, has a bonus scheme explicitly tied to AI-driven output. Automate two days of work a week, and the cheque arrives in December. Sounds like a win-win.
Except the maths doesn’t work for the human. “I do not receive two days off or a 40% pay rise,” he tells me. “The higher output simply becomes the new baseline.” In the short term, the productivity spike might secure a promotion. Long term? He has effectively mapped the exact coordinates of his own redundancy. He now views himself less as a specialist and more as a manager — managing people, yes, but also “managing AI agents, or AI loops, or whatever you want to call it.”
The Two-Tier Workforce
It isn’t just about bonuses. It is about survival. Pamela — a US-based senior executive at a major consultancy who asked for her name to be changed — describes a “quiet” revolution. No formal mandate exists, but the performance reviews tell the real story. “You’ve got to demonstrate fluency and fluidity as one of your key achievements,” she says. “AI fluency beats credentials every day.”
The result is a brutal stratification. Two employees, same title, same tenure. One treats AI as a threat, the other as a power tool. The gap widens exponentially the moment leadership spots it. “Somebody that has 15, 20 years of experience and no AI fluency will be passed over for those that have, say, three years, but are fast with the tools,” Pamela warns. “If you’re not visibly using AI, you see slower promotion timelines. It’s harder to be seen, and it’s harder to fight being on that shortlist.”
Legal, But Is It Fair?
Employment lawyers confirm the goalposts can legally be moved. Tina Chander, partner at Weightmans, says bosses are “in the clear” legally, but the fairness questions are stacking up. If AI makes a worker 40% more efficient, does the employer simply raise the target by 40%? “Is that fair?” Chander asks. “And if an employee is effectively doing more because AI has made them more efficient, should they be rewarded differently? Or are they effectively making themselves redundant, thus acting as a disincentive to be productive?”
The legal landscape is shifting, too. From January 2027, UK employees will have six months to submit unfair dismissal claims — double the current window — and they’ll qualify after just six months of service, down from two years. HR consultant Tina Rahman of HR Habitat argues employers are failing the transparency test. They want the productivity gains and the reduced outsourcing costs, but they aren’t articulating the “end purpose” to the workforce. “Because they misunderstand it, this is not being reflected to employees,” she says.
Performative Compliance
Kamila Miller, an applied AI researcher at Henley Business School, has a warning for the KPI-obsessed. Make AI usage a Key Performance Indicator, and you get exactly what you measure: theatre. “People will log their interactions to hit the metric, route work through a chatbot that did not need it, and generate AI-flavoured outputs that look productive on a dashboard,” she says. “You will measure adoption. You will not measure judgement, learning, or better decisions. You have not made people more skilled — you have made them more obedient.”
Some giants are already course-correcting. Duolingo’s Luis von Ahn admitted in April that staff were asking, “Do you want us to use AI for AI’s sake?” The company dropped AI from performance reviews, refocusing on output quality regardless of the tool. Amazon followed in May, shuttering an internal leaderboard after staff gamed it by feeding the models needless busywork.
The Endgame
For now, the leaderboard remains the law of the land for many. Pamela, in her mid-50s and a decade from retirement, is visibly leaning into the tools to secure the pension and health benefits she can’t afford to lose. Trevithick is hedging differently: building side projects and assets where he owns the AI leverage, not the employer. “If AI can do a job better than you can, it makes sense for the business to replace you,” he says. “That’s how the capitalist model works. So it’s about, how can you move into a position where you own assets where you can leverage AI, and then you benefit?”
The game has changed. The only question is whether you are playing to win, or playing to stay employed.
Why it Matters
We are witnessing the real-time restructuring of the white-collar social contract. When AI fluency becomes the primary currency for career advancement, experience becomes a depreciating asset and the incentive structure flips: workers are rewarded not for judgement or creativity, but for the speed with which they automate their own tasks. This creates a workforce optimised for obedience over insight, storing up a future where institutional memory is erased and the only metric that matters is how fast you can prompt the machine. If businesses don’t decouple adoption from value creation — and fast — they won’t just lose trust; they’ll build organisations that are efficient at doing the wrong things, perfectly.
