The conversation about artificial intelligence and employment has grown increasingly polarised. One side warns of mass job displacement; the other points to new opportunities. But the real story, according to venture capitalist Jon McNeill, may sit in the unglamorous middle: the vast infrastructure required to sustain AI systems.
McNeill, former president of Tesla and chief operating officer of Lyft, brings credibility to this argument. His track record includes scaling companies through periods of explosive growth. When he talks about AI infrastructure jobs, he is not speculating about future potential; he is observing present reality.
The scale of current investment is genuinely remarkable. Global AI infrastructure spending is expected to reach between $400 billion and $450 billion in 2026, covering new data centres, semiconductor plants, and power-grid expansion. This is not venture capital betting on distant applications. This is capital flowing into concrete, steel, and electrical systems.
The employment consequence is measurable. In 2026, the expansion of AI infrastructure is a major driver for construction employment, with large-scale data centre projects generating thousands of jobs and stimulating local economies across the United States. In specific cases, the numbers are substantial. One company's campuses in New Mexico and Wisconsin are each projected to generate approximately four thousand construction positions, with many anticipated to be filled by local union members, while its Texas location has already utilised over eight thousand construction workers since work started.
But infrastructure jobs are only part of the picture. Demand is highest for project managers, superintendents, and electrical specialists with experience in high-density, AI-ready data centres, with wages in these sectors having increased by as much as 40 to 60 per cent since 2024. Once facilities are operational, they require ongoing technical maintenance. Microsoft has provided ongoing employment to thousands of construction workers for almost two decades across its datacentre regions, with hundreds of technicians enjoying permanent jobs in those datacentres, earning salaries well above the median income for local areas.
The broader argument McNeill appears to be making is about scale. He has observed that AI infrastructure development will operate on a decades-long timeline, much like the internet buildout that began in 1997. Sustained demand for infrastructure specialists, cooling engineers, electrical workers, and facility managers creates genuine labour market opportunity.
Yet the employment picture remains complicated. There is a mildly negative correlation between employment trends and AI usage, suggesting that AI may be depressing job growth, a trend especially evident in certain tech industries, including cloud, web search and computer systems design, with these three industries stopping growth at the end of 2022, just after the release of ChatGPT. AI is eroding the bottom rungs of career ladders by automating many intellectually mundane tasks that junior employees typically handle, with workers aged 22 to 25 in AI-exposed fields experiencing a 13 per cent relative decline in employment even as older colleagues saw gains in the same sectors.
The divergence is striking. Entry-level knowledge workers in tech, consulting, and creative fields face displacement. Meanwhile, skilled trades and infrastructure specialists are in tight supply and commanding premium wages. Analysts estimate the global industry will need more than 200,000 additional electricians, technicians, and project managers by 2026.
This means the policy challenge ahead is not whether AI creates jobs overall. The infrastructure expansion makes clear it does. The challenge is whether displaced knowledge workers can transition to infrastructure-related roles, and whether education and training systems can prepare the workforce for specialised technical work in data centre development and maintenance.
McNeill's perspective offers neither reassurance nor alarm. He is describing what he observes: a decade-long infrastructure project that will generate substantial employment alongside technological displacement. The labour market, if current trends hold, will reward specialisation and technical skill while penalising routine work. That may be the defining employment reality of the next several years, regardless of broader AI adoption.