We are looking at AI through the wrong lens. While the corporate world debates whether AI is a bubble or a tool for headcount reduction, they are missing the actual revolution happening on the ground.
A recent New Yorker article by Joshua Rothman highlights this perfectly: a seven-year-old boy used AI to go from zero basic coding skills to programming complex video games in weeks.
This isn’t just automation. It is the radical acceleration of human capital.
Economist Theodore Schultz defined human capital as the investment we make in ourselves to improve the quality of our effort. Historically, this was a slow climb. But AI allows us to leapfrog learning curves, acting as a multiplier that makes us dramatically more capable almost overnight.
Here is the trap: Most companies are structurally blind to this multiplier.
The traditional ideal hire builds skills feverishly before the job, then settles into a stable “cog in the machine”. But AI prevents that settling. It allows employees to continuously expand their capabilities - writing software, analyzing complex research, or diagnosing problems they couldn’t handle yesterday.
When a worker’s skills outpace their role, the job becomes a prison.
Two ways companies will lose their best people:
1. Capturing value without sharing it
If an employee uses AI to become significantly more capable, but compensation stays flat, the firm captures the return on that individual’s personal investment. Rational in the short term. But the employee notices the asymmetry - and starts looking elsewhere.
2. Structural dead-ends
Money isn’t the only issue. Employees who expand their capabilities need room to run. If a QA analyst uses AI to build automation frameworks and own the CI/CD testing pipeline, keeping them in manual test execution wastes their expanded capacity - and signals there’s no path forward. They’ll find one somewhere else.
Until companies figure out how to value this cognitive boost and create pathways for expanded capabilities, high-performing workers have one option: take their portable, AI-enhanced skills to a competitor who understands that the org chart and compensation should shift around the talent, not the other way around.
If you treat your AI-enhanced employees like cogs, don’t be surprised when the machine breaks.