AI is collapsing the talent cycle from years to weeks, and the supply gap is already acute
The mobile engineering shortage took roughly two years to resolve. The current scramble for AI-capable talent is playing out in weeks, and the institutional mechanisms that historically absorbed such mismatches are not keeping pace.
Adam Ward draws the comparison in terms that make the timeline compression concrete. The mobile engineering shortage, he argues, ran about two years before boot camps and retraining programs caught up with demand. The current moment, he says, is playing out in weeks. That is not a modest acceleration. It means the institutional mechanisms that historically absorbed talent mismatches, retraining curricula, certification programs, university pipelines, are moving at a pace the market is not waiting for.
Ward also flags a supply-demand imbalance that is already acute for certain roles. He does not quantify it, but the structural read is plain: demand arrived before supply had any chance to form, and the gap is not closing on a timeline that hiring managers find comfortable.
Matt Swulinski puts a harder figure on the scarcity. Fewer than one percent of job candidates, in his estimate, possess the depth needed to operate effectively in this environment. That is a striking number, and it is worth holding loosely: it is his read on the candidate pool he encounters, not a census. But it rhymes with the directional picture that Ward’s framing implies: organizations need people who do not yet exist in sufficient numbers, and no near-term pipeline is producing them fast enough.
What makes this moment genuinely unusual is not just the shortage but the speed. Talent shortages are not new. What is new is the rate at which new role categories are crystallizing and the rate at which organizations need to fill them. When the mobile moment played out, employers had time to wait for the ecosystem to produce people. That patience is no longer available at the same scale.
We do not expect you to have a background in AI because again, this is evolving so quickly, but we definitely expect high curiosity about it. Ian Silber
One organizational response is to stop waiting for supply and instead bet on aptitude. Ian Silber, discussing design hiring at OpenAI, makes the logic explicit: “We do not expect you to have a background in AI because again, this is evolving so quickly, but we definitely expect high curiosity about it.” That framing is worth noting carefully. It is not an argument that credentials are irrelevant. It is an argument that, in a domain moving this fast, curiosity and learning speed may predict performance better than a background that could itself be obsolete within a year or two.
That approach carries a real tradeoff. Hiring for aptitude over existing skill shifts the training burden onto the organization. It works when the employer has the infrastructure and the time to develop people. It does not scale well when every organization in the market is making the same bet on the same thin pool of high-curiosity generalists.
The broader dynamic this points toward is a structural mismatch between the speed at which AI is creating new organizational needs and the speed at which labor markets can respond. Historically, labor markets are slow. They reprice gradually, they retrain over years, and they build new supply through educational institutions that operate on decade-long cycles. None of those mechanisms are calibrated for a shift happening in weeks.
Whether the imbalance resolves through rapid on-the-job development, through organizations narrowing their ambitions to match available talent, or through some combination, is not yet clear. What Ward’s framing suggests, and what Swulinski’s figure implies, is that the resolution is not imminent. The pipeline is not catching up. Organizations that are building around AI-driven workflows are, for now, competing over a very small number of people, and the competition is only getting louder.