AI has created more than 750,000 new jobs in the U.S. since 2023, according to LinkedIn estimates reported by The Wall Street Journal in September 2026. Most of them are not the $180,000 engineering roles that make headlines. The biggest group is data annotators, followed by data center workers and AI engineers.
At the same time, U.S. employers have cited AI in 120,136 announced job cuts so far in 2026, more than any other reason, and the share of Americans who call AI’s risks very serious jumped from 35% to 42% in a single month. Below are the numbers on both sides, with the source and date for every figure.
Key takeaways
- AI-linked roles account for more than 750,000 new U.S. jobs since 2023 (LinkedIn estimates, reported September 2026).
- Data annotators lead with about 282,000 new positions, followed by about 117,000 data center jobs and 105,000 AI engineers.
- The median posted salary for AI-related roles on LinkedIn is about $180,000, compared with $80,000 across all jobs.
- Employers have cited AI in 120,136 announced U.S. job cuts through September 2026, about 21% of all cuts (Challenger, Gray & Christmas).
- 42% of U.S. adults now say the risks of AI causing harm are very serious, up from 35% in late August 2026 (Economist/YouGov).
- 47% of U.S. adults also say AI will have many positive benefits, versus 31% who say it won’t.
How many jobs has AI created in the U.S.?
LinkedIn estimates that AI was behind more than 750,000 new U.S. jobs from 2023 onward. The figure was published by The Wall Street Journal on September 23, 2026, in its report on the AI infrastructure build-out, and LinkedIn’s role breakdown was widely shared in early October 2026.
Five roles make up roughly 548,000 of the total. The rest is spread across other AI-linked job titles.
| AI-linked role | Approximate new U.S. jobs | What the work involves |
|---|---|---|
| Data annotators | 282,000 | Labeling and checking the data AI models learn from; often contract work |
| Data center jobs | 117,000 | Running and maintaining the facilities that house AI hardware |
| AI engineers | 105,000 | Building, deploying and maintaining AI models and AI-powered software |
| Heads of AI | 29,000 | Leading a company’s AI strategy and teams |
| Forward-deployed engineers | 15,000 | Working on site with customers to turn AI tools into working systems |
| Total, all AI-linked roles | 750,000+ | Includes roles beyond the five above |
Globally, LinkedIn counts even more. Writing for the World Economic Forum in January 2026, LinkedIn COO Dan Shapero said AI had created 1.3 million new roles worldwide in two years, such as AI engineers, forward-deployed engineers and data annotators, plus more than 600,000 new AI-enabled data center jobs. He also reported:
- AI Engineer is one of the fastest-growing jobs on LinkedIn over the past three years.
- U.S. roles that require AI literacy rose 70% year over year.
- LinkedIn’s data shows AI is not the main reason hiring has slowed. Economic uncertainty, monetary policy and post-pandemic rebalancing matter more.
Our AI engineer job description covers the duties and skills behind the fastest-growing of these roles.
How much do AI jobs pay?
AI-related roles on LinkedIn have a median posted salary of about $180,000, more than double the $80,000 median across all jobs. That figure hides a wide spread, though. The largest new AI role, data annotator, has a median listed pay of about $51,000, below the median for jobs overall.
So when people say AI jobs pay $180,000, that is true of the engineering and leadership end. The most common new AI job pays far less, and much of it is contract work.
How many jobs has AI cut?
Layoff announcements that name AI have climbed fast in 2026. Outplacement firm Challenger, Gray & Christmas reported on October 1, 2026, that AI has been cited in 120,136 announced U.S. job cuts so far this year. That is about 21% of all announced cuts and makes AI the leading reason year to date.
| Challenger metric (2026) | Figure |
|---|---|
| Total announced U.S. job cuts, January to September | 573,195 (down 39% from 2025) |
| Cuts citing AI, January to September | 120,136 (about 21% of the total) |
| Cuts citing AI in September alone | 3,961 (about 9%, fifth-most cited reason) |
| Technology sector cuts, January to September | 165,925 (up 54% from 2025) |
| Announced hiring plans, January to September | 210,612 (up 3% from 2025) |
Two things are worth keeping in mind. First, these are announced cuts, not confirmed layoffs, and the reason is whatever the employer states. Second, total job cuts are well below last year, so AI’s share is rising partly because other reasons have fallen.
The broader job market is soft rather than collapsing. The Bureau of Labor Statistics reported on October 2, 2026, that U.S. payrolls rose 29,000 in September, after an average monthly gain of 45,000 over the prior 12 months. Unemployment was 4.2%. For long-range forecasts of how many roles AI could displace, see how many jobs AI will replace by 2030.
Are Americans worried about AI?
Yes, and worry is rising quickly. The Economist/YouGov poll of U.S. adult citizens, taken September 18 to 21, 2026, asked how serious the risks are of AI causing harm:
| How serious are the risks of AI causing harm? | Share of U.S. adults |
|---|---|
| Very serious | 42% |
| Somewhat serious | 31% |
| Slightly serious | 20% |
| Not at all serious | 7% |
The share saying very serious rose from 35% in the August 28 to 31 poll. Most of that jump came from Democrats, and Republicans barely moved:
Concern does not mean people expect nothing good from AI. In the same poll:
- 47% of U.S. adults say AI will have many positive benefits, while 31% say it won’t.
- Among those who call AI’s risks only slightly serious, 69% expect many benefits.
- In open-ended answers, the most common worries were loss of human control, job loss and harm to humans.
- The most commonly expected benefits were medical and health care advances, higher productivity, and scientific and technological progress.
Other 2026 surveys point the same way, especially on jobs:
| Survey finding | 2026 | Earlier | Source |
|---|---|---|---|
| Workers who worry technology could make their job obsolete | 27% | 20% (2025), 13% (2017) | Gallup, Aug 2026 |
| Same worry, workers aged 18 to 44 | 34% | n/a | Gallup, Aug 2026 |
| Same worry, workers aged 45 and older | 19% | n/a | Gallup, Aug 2026 |
| Americans who expect AI to reduce U.S. jobs over 10 years | 79% | 73% (2025) | Bentley-Gallup, Jul 2026 |
| Americans who say AI does more harm than good | 39% | 31% (2025) | Bentley-Gallup, Jul 2026 |
Are younger people less worried about AI?
It is a common assumption that younger workers, who grew up with the technology, are more relaxed about AI. The 2026 survey data does not support it.
- In the Economist/YouGov crosstabs for September 18 to 21 (registered voters), 48% of 18 to 29 year olds called AI risks very serious, tied for the highest of any age group.
- In Bentley-Gallup’s 2026 survey, 47% of adults aged 18 to 29 said AI does more harm than good, up from 36% in 2025. That was the largest jump of any age group.
- Gallup found that workers under 45 are now more likely to worry about technology making their jobs obsolete than about layoffs, wage cuts or reduced hours.
The most optimistic group is people aged 30 to 44. In the same crosstabs, 61% of them said AI will have many positive benefits, the highest of any age group.
Where are AI jobs being created?
Much of the new hiring follows the physical build-out. A September 2026 paper by economist Stijn Van Nieuwerburgh, published by the Brookings Institution, projects $10.3 trillion of U.S. investment in data centers and related AI infrastructure from 2025 to 2032. That averages about 3.6% of GDP a year, a larger share of the economy than the railroad or highway build-outs.
That money turns into jobs for electricians, mechanical and controls engineers, commissioning teams and critical facilities staff, and they are concentrated in a few metros. Our US data center map tracks where the projects are, and the data center statistics hub covers capacity, power and employment.
Local opposition is a real constraint. An Economist/YouGov poll from August 28 to 31, 2026, found 63% of Americans would oppose a new data center being built in their community, and 60% expect one would raise their electricity costs.
What do these AI jobs statistics mean for employers?
For companies hiring technical talent, a few things stand out:
- The pay premium is real at the top. AI-related roles post a median salary of about $180,000, and budgets for AI engineers and AI leaders need to reflect that.
- The physical build-out is the bigger hiring challenge. Data center and power projects need experienced electrical, mechanical and controls engineers, and those people take years to train. Our data center workforce shortage statistics show how tight that pool already is.
- AI skills are becoming a baseline. With U.S. roles requiring AI literacy up 70% in a year, job descriptions for engineers and IT staff increasingly ask for AI experience alongside core skills.
- Candidates are uneasy. With worry about AI rising among workers under 45, being clear about how a role uses AI, and how it doesn’t, can help close offers.
Recruiter tip: The hardest AI-era hires are rarely data annotators. They are the engineers who build and run AI infrastructure: electrical and controls engineers on data center projects, and software engineers who can put AI tools into production. Start those searches early, because the best candidates usually have several offers.
Apollo Technical recruits for these roles. See our data center recruiting, controls engineering recruiters, electrical engineering recruiters and IT staffing pages, or read our guide on how to hire data center engineers.
Frequently asked questions
How many jobs has AI created?
LinkedIn estimates AI-linked roles account for more than 750,000 new U.S. jobs since 2023, led by about 282,000 data annotators, 117,000 data center jobs and 105,000 AI engineers. Worldwide, LinkedIn counted 1.3 million new AI-related roles in two years as of January 2026.
Is AI creating more jobs than it is eliminating?
There is no single answer yet, because the two numbers are measured differently. LinkedIn’s 750,000 figure is an estimate of new roles since 2023. Challenger’s 120,136 figure counts announced cuts in 2026 where the employer named AI. Neither is a full count, and the overall U.S. job market has slowed for several reasons, not only AI.
How much do AI jobs pay?
The median posted salary for AI-related roles on LinkedIn is about $180,000, compared with $80,000 for all jobs. Pay varies widely by role. Data annotators, the largest group of new AI jobs, have a median listed pay of about $51,000.
What percentage of Americans are worried about AI?
In the September 18 to 21, 2026, Economist/YouGov poll, 42% of U.S. adults said the risks of AI causing harm are very serious and another 31% said somewhat serious. Only 7% said not at all serious.
Are AI layoffs increasing?
AI is the most-cited reason for announced U.S. job cuts in 2026, named in 120,136 cuts through September, according to Challenger, Gray & Christmas. Total announced cuts are down 39% from 2025, though, so AI’s share is growing as other reasons fall.
Sources
- The Wall Street Journal, “The AI Build-Out Is Becoming the Biggest Economic Bet in U.S. History”, September 23, 2026 (LinkedIn estimates)
- World Economic Forum, “AI has already added 1.3 million jobs, LinkedIn data says”, January 15, 2026
- Challenger, Gray & Christmas, September 2026 Job Cuts Report, October 1, 2026
- U.S. Bureau of Labor Statistics, The Employment Situation, September 2026, October 2, 2026
- Economist/YouGov poll, September 18 to 21, 2026 and crosstabs
- Economist/YouGov poll on data centers, August 28 to 31, 2026
- Gallup, “More U.S. Workers Fear Losing Their Jobs to Technology”, September 15, 2026
- Bentley University-Gallup, “Americans Cool Toward AI”, July 28, 2026
- Brookings Institution, “Financing the AI buildout”, September 2026
