AI Anxiety Hits Early Careers: Pichai, Gates and the Data Behind It

Young workers aged 22 to 25 are falling behind in the occupations most exposed to artificial intelligence, according to a Stanford Digital Economy Lab study updated in August 2026. The damage, Stanford found, shows up not in layoffs but in jobs that never materialised — entry-level positions that were simply never created.

The data point lands amid a run of blunt statements from the people building the technology. Sundar Pichai, CEO of Google, told the BBC in an interview on 18 November 2025 that an AI agent could eventually do his job. "I think what a CEO does is maybe one of the easier things for AI to do one day," he said.

Bill Gates went further. In an interview with The Observer, published in late October 2026, he described AI as an "evolutionary event" in which humans may no longer be the most intelligent beings on the planet. His workplace prediction covers not only routine tasks but skills of invention and white-collar work, with humanoid robots eventually extending automation into physical labour. His summary of the employer incentive is blunt: "Hire the robot, fire the human and don't look back."

What do the leaders actually predict?

Elon Musk, in a July 2026 interview with The Economist, suggested AI could soon perform almost any digital job better than humans, with employment becoming optional — people working by choice rather than necessity. His vision of abundance, however, leaves open the question of who owns the wealth machines generate, and who is left struggling to find a place in the new economy.

The payroll evidence remains more contained than the rhetoric. Stanford found no evidence of widespread, economy-wide job displacement associated with AI. The frequently cited World Economic Forum forecast — 92 million jobs displaced and 170 million created by 2030 — covers technological change, demographic shifts and the green transition together. It is not an AI-only figure, and readers should treat it as such.

What does this mean for your market position?

For an individual worker, the pressure operates on two fronts. The first is hiring: if AI writes the first draft, prepares the first brief and produces the first analysis, the training ground for tomorrow's accountants, lawyers and managers narrows. The second is pay. The Economist, in an article titled Ignore AI and Ask for That Pay Rise, raises the concern that fear of replacement may itself suppress wages — a worker who believes a chatbot is waiting for their seat does not ask for a raise.

The historical parallel is imperfect but instructive. ATMs spread in the 1980s, and many predicted the end of the bank teller. They were only partly right: teller numbers per branch fell, but lower operating costs let banks open more branches. AI differs in kind, not just degree — it writes, reasons, analyses and makes decisions, abilities once considered uniquely human.

What remedies are on the table?

Gates has proposed reserving certain jobs for humans where human involvement is essential, and has suggested governments consider taxing AI use, including the tokens businesses purchase to run AI systems. The article's policy argument extends further: governments must share the gains through retraining and social protection, and employers must find new ways to give young people the experience once gained through entry-level roles.

For workers, the prescription is competence rather than optimism. AI literacy — using the tools, questioning their outputs, understanding their limits — is positioned alongside critical thinking, creativity and judgement. What is not said, notably, is how quickly any of these remedies can be built; Stanford's own data on the missing entry-level cohort suggests the hiring gap is already open.

The uncertainty itself is now a labour-market force. By the time regulation, corporate adaptation and education catch up, the technology may already have moved on.