AI Was Supposed to Destroy Jobs. Instead, 1 Million New US Roles Have Emerged: But Who Gets Them?

Artificial intelligence might eventually leave many people without work, but there is little evidence of an economy-wide AI jobs crisis so far.

On 4 September, the Bureau of Labour Statistics reported that the US economy added 162,000 jobs in August, far above expectations, while the unemployment rate remained at 4.1%.

US Labour Market Defies AI Fears

At 4.1%, the jobless rate sits below levels seen in nearly nine out of ten months across the last fifty years. According to The Economist, fresh graduates and other youth—frequently painted as the initial casualties of automation—are faring exceptionally well, with the margin separating youth unemployment from the broader rate hovering near a multi-decade low.

While some firms and employees face major disruption from technology, hiring within professional and business services currently tracks roughly 10% under its 2015-19 average. At the same time, major tech corporations such as Meta and Microsoft are cutting staff while restructuring their businesses amid heavy investment in AI.

Firms such as Block, which operates Square and Cash App, along with Intuit, the developer behind TurboTax and QuickBooks, are increasingly using AI agents and software to automate tasks previously performed by people. Challenger, Gray & Christmas, an employment consultancy, estimates that US companies have announced roughly 16,000 AI-related job cuts a month on average so far this year.

AI Boom Drives Data-Centre Hiring

Those job cuts are relatively small compared with the wider churn in the US labour market, where businesses record around 1.7 million layoffs and discharges each month. Yet automation is already generating fresh roles to make up for the ones it removes.

Massive spending on data centres and electricity has triggered a scramble for builders and infrastructure experts.

New tech firms are expanding rapidly, while established companies build out dedicated AI teams to stay competitive. By driving up individual efficiency, the technology may even be fuelling a higher demand for human labour.

The Economist estimates that the AI boom has been associated with roughly one million new US roles, easily outpacing the estimated 200,000 layoffs attributed to AI since mid-2023.

This expansion stems largely from a massive infrastructure boom. Goldman Sachs estimates that annual spending on chips, servers, data centres, cooling systems and power infrastructure needed to run AI is roughly $500 billion (£369.94 billion) above 2022 levels.

Census Bureau data shows data-centre construction running at an annual rate above $75 billion (£55.49 billion), up nearly 60% from a year earlier, creating heavy demand for electricians, HVAC technicians, grid engineers and equipment installers.

The Economist tracked five data-centre industries, finding that employment has risen by roughly 320,000 above trend since 2023. The Bureau of Labour Statistics projects utilities as the fastest-growing sector through 2035.

While grid upgrades contribute, AI appears to be a major driver of this expansion; data-centre vacancies have doubled in two years despite falling overall postings, and LinkedIn estimates that nearly half a million US data-centre jobs were created between 2023 and 2025.

Data-Centre Jobs Offer Higher Pay

Data-centre installation and maintenance roles pay about 40% more than comparable positions elsewhere, according to Indeed. Official statistics reflect this trend; in the year to June, average hourly earnings rose by more than 13% in electrical-equipment manufacturing and nearly 8% among electrical contractors.

Despite higher pay, labour shortages persist. Donald Leavens of the National Electrical Manufacturers Association has highlighted the difficulty companies face in finding workers for the expanding electrical infrastructure industry.

AI Creates a New Wave of White-Collar Roles

AI is also generating white-collar roles like model engineers, data annotators, forward-deployed engineers and heads of AI. LinkedIn reports that postings for these positions have roughly doubled since 2023-24.

Burning Glass Institute chief economist Gad Levanon estimates that AI-specific roles now account for roughly 1% of professional positions, translating to about one million jobs in America, with concentrations reaching 4-5% in tech and life sciences. LinkedIn similarly identified about 640,000 AI-related jobs in the US between 2023 and 2025.

Comparing 2022 growth to overall professional trends, The Economist found that jobs close to the AI boom—including software developers, engineers, mathematicians and data scientists—have gained roughly 730,000 positions above normal expectations. While AI didn't trigger every single one of those roles, it appears to have driven a significant share of them.

AI Productivity Could Create More Jobs

Productivity boosts represent a third source of employment. When automation enables attorneys to write agreements more quickly or lets researchers analyse financial documents in minutes, lower costs for these professional services can spark enough extra demand to generate additional jobs overall.

Occupations deemed vulnerable to AI are instead growing. Between 2023 and 2025, paralegal employment rose 11%, and market-research analysts grew 6%, outpacing the 2.5% national average. BLS forecasts project continued rapid growth in professional services driven by demand for AI systems and consulting.

Routine Jobs Are Already Taking a Hit

Certain fields are taking a direct hit. Since January 2023, customer-service staffing has dropped by around 10%, while roles for secretaries and administrative assistants have fallen by roughly 15%. These areas rely heavily on routine tasks where software agents now excel. Consequently, the Bureau of Labor Statistics anticipates office and administrative support sectors will lose 752,000 positions by 2035.

As AI expands, it will generate new roles. History shows past tech waves have also produced entirely new categories of work. More than 6 million Americans now work in computer-related occupations that did not exist before the computer era, while another 8 million work in newer industries such as e-commerce, the gig economy and content creation.

Future workers might manage autonomous agents or resolve disputes between them, much like influencers emerged two decades ago.

Frequently Asked Questions

• How is AI affecting job creation in the US?
  AI is creating new roles in data centers and white-collar positions while also causing job cuts in routine tasks.
• What sectors are seeing job growth due to AI?
  Data centers, electrical infrastructure, and AI-specific roles like model engineers and data annotators are seeing growth.
• Which jobs are most vulnerable to AI automation?
  Routine jobs such as customer service and administrative roles are most vulnerable to AI automation.