Fastest-Growing AI Jobs and the New Roles AI Is Creating

The fastest-growing AI jobs include AI engineer, which topped LinkedIn's Jobs on the Rise 2026 list for the US, plus data scientists and computer and information research scientists, which the BLS projects to grow 34.6 and 21.8 percent from 2025 to 2035. Newer titles such as forward-deployed engineer, evals engineer and AI governance specialist are growing from a small base.

Definition: A fast-growing AI job is a role whose postings or employment are rising quickly because companies are building, deploying, governing or powering AI systems. Some are established jobs with new demand; others are titles that barely existed a few years ago.

Fast growth and lots of jobs are not the same thing. A title can triple and still be rare.

Fastest-growing AI jobs: what the growth data says

Every source we checked shows AI demand rising. They disagree on how big it is, because each one counts something different.

Source	What it measures	Latest reading
Stanford HAI AI Index 2026 (Lightcast data)	US postings that ask for AI skills	2.56% of postings in 2025, up from the year before
Indeed Hiring Lab AI tracker	US postings mentioning AI or machine learning terms	6.74% on August 31, 2026, versus 3.44% a year earlier
PwC 2026 Global AI Jobs Barometer	Jobs requiring specific AI skills, 27 countries and territories	Growing 69%, versus 9% for the overall jobs market
BLS Employment Projections 2025 to 2035	Projected US employment by occupation	Computer and mathematical jobs +7.3%, against +3.5% for all jobs

A few things worth noticing:

The skills are spreading faster than the job titles. The AI Index chapter, using Lightcast data, found "agentic AI" appeared in about 0.9 percent of US AI job postings in 2024 and about 3.3 percent in 2025. That is close to four times the share in one year, still from a small base.
AI is leaking into non-AI jobs. Indeed's measure is broader than Lightcast's because it counts any posting that mentions AI terms, including marketing, finance and operations roles, which likely explains much of the gap.
Pay is moving with demand. PwC's 2026 barometer puts the average wage premium for workers with AI skills at 62 percent, up from 57 percent the year before. Carta's State of Startup Compensation 2026 found median salaries for AI engineers at startups on its platform rose 10 percent over the prior 18 months, while salaries for all startup employees rose 5.1 percent since January 2024, per its May 2026 report.

Our read: the direction is clear. The precise size depends on the definition, so we'd be wary of any single "AI jobs grew X percent" headline.

The fastest-growing AI jobs right now

These are established roles where demand is measurably rising.

AI engineer (often listed as machine learning engineer). LinkedIn's Jobs on the Rise 2026 report, which ranks US titles by growth over the past three years, put AI engineer at the top. The job is mostly building products on top of models: connecting them to data, wiring up agents, testing and shipping. Many people in it came from ordinary software engineering. Big employers are leaning in: Gergely Orosz and Jessica Salmon of The Pragmatic Engineer reported in May 2026, using TrueUp data, that many larger tech companies had 50 to 100 percent more AI engineering listings than a year earlier. Our guide on how to become an AI engineer covers the skills.

Data scientist. The BLS projects 34.6 percent growth from 2025 to 2035, the third-fastest of any US occupation in its latest projections. The BLS links the strength of the computer and mathematical group to "the continued proliferation of digital tools and AI solutions" across the economy.

Computer and information research scientist. Projected at 21.8 percent, also in the BLS top ten. It is a small occupation with a high bar; the BLS lists a master's degree as the typical entry point.

AI consultant or strategist. Also on LinkedIn's 2026 list. These people help companies decide where AI fits and run the rollout. Domain knowledge (health care, finance, logistics) often matters as much as technical depth.

Data annotator. LinkedIn's list included data annotators too: people who label data and grade model answers, often on contract, with volume that can swing.

AI infrastructure and data center roles. Not glamorous, but growing. LinkedIn's 2026 list included data center technicians, and the BLS expects electricity demand to rise over its projection period "due in large part to AI adoption" and related infrastructure such as data centers.

The new roles AI is creating

These titles were rare a few years ago. Most are still small, which can help: fewer people have the experience, so the bar is lower.

Anthropic's public job board, for one, has carried titles in 2026 that map to most of the roles below, from forward-deployed engineer to model evaluations, AgentOps and safeguards. Titles vary by company, so read the job description, not the label.

Forward-deployed engineer

An engineer who works inside a customer's company to get an AI product running in their real systems. Part software engineer, part consultant. It suits people who like building and like customers, and it often rewards fast prototyping over deep research. The logic is simple: a model that never gets deployed earns nothing. a16z's Joe Schmidt made the case in 2025 that this hands-on implementation work is how AI startups win complex enterprise customers, and noted that OpenAI and Anthropic hire for it too.

Evals engineer or evaluation specialist

Someone who designs the tests that tell a team whether a model or agent is good enough to ship: test cases, scoring, catching regressions. It is one of the easier AI skills to demonstrate on your own. Build an evaluation set for a task you know well and publish the results.

AI trainer and expert data specialist

A more senior version of annotation. Lawyers, doctors, accountants and engineers write the hard examples and grade the answers that teach models a domain. For a professional considering an AI career, this can be the most direct bridge.

Agent operations

As companies deploy AI agents that take actions (filing tickets, updating records, sending emails), someone has to configure, monitor and fix them. Titles vary: AgentOps engineer, AI operations, automation lead. In our view, this is a natural next step for operations, RevOps and support people who already own a workflow.

AI governance, safety and policy

Companies need people who can set rules for how AI is used, assess risk and keep up with regulation. The IAPP's AI Governance Profession Report 2025, based on a spring 2024 survey of 670+ professionals, found 77 percent of organizations were working on AI governance, and only 1.5 percent said they would not need more staff for it within a year. Those responsibilities most often sat in privacy (22 percent) or legal and compliance (22 percent) teams. Labs hire safety analysts in this family too.

If you don't code, our guide to non-technical AI jobs goes deeper on the governance, operations and go-to-market side.

What jobs will AI create, and which will it replace?

This is where we'd hold forecasts most loosely. Here's what the better sources say.

The global employer view. The World Economic Forum's Future of Jobs Report 2025, a survey of 1,000+ employers representing over 14 million workers, expects 170 million jobs to be created and 92 million displaced from 2025 to 2030, a net gain of 78 million. AI and machine learning specialists rank third among its fastest-growing roles, behind big data specialists and fintech engineers. Clerical roles such as cashiers, administrative assistants and bank tellers lead its declines.

The US government view. The BLS's 2025 to 2035 projections expect AI-powered automation to reduce demand for several office and administrative support occupations, and generative AI to limit demand for some jobs in arts, design, media and entertainment. In a February 2025 article, BLS economists projected software developer employment to grow 17.9 percent from 2023 to 2033, partly because AI productivity may lower prices and raise demand for software, while customer service representatives were projected to decline 5.0 percent.

The early-career warning. The August 2026 update of the "Canaries in the Coal Mine" paper by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen, using ADP payroll data through June 2026, found employment of 22 to 25 year olds in the most AI-exposed occupations stood 19 percent below where it would have been had it kept pace with less-exposed peers. Declines concentrated where AI substitutes for tasks; where it complements workers, employment was flat or rising. The authors present these as descriptive facts, not proof of cause.

The twist. PwC's 2026 barometer found entry-level roles that now ask for more senior skills grew 35 percent since 2019, while other entry-level roles shrank 10 percent. Entry-level jobs aren't vanishing so much as changing shape.

Putting it together, our view is that AI is likely to create more specialized jobs than it removes over the long run, but the jobs it removes and the jobs it creates are often held by different people. The adjustment may be hardest at the bottom rung.

Best AI careers to start in 2027: our picks

No list can predict the future, so here are our criteria first. The test: does the job complement AI rather than compete with it? Does demand come from deploying AI, not just from research? Can you show proof of skill without a PhD?

Roles that pass, in our view:

AI engineer, for people who code. The largest technical door, with clear proof (shipped apps).
Forward-deployed or solutions engineer, for builders who like customers.
Evals and quality, for detail-oriented people with a technical or domain background.
Agent operations, for operators who already own a messy workflow.
AI governance and risk, for people from privacy, legal, compliance or security.
Data science with a domain, for analysts in health care, finance or another field AI is entering.

If you're choosing between these, our guide to AI career paths walks through what each looks like day to day, and our list of best entry-level AI jobs covers the starting rungs.

"AI jobs are a bubble"

The skeptic's case is reasonable. Hiring follows funding, and AI funding is concentrated. A few model companies could consolidate, and titles invented in a hot market can disappear in a cold one. Some forward-deployed and annotation roles may get automated themselves.

But.

Most of the growth above comes from companies deploying AI in their own work, not from AI labs. Banks, insurers, hospitals and retailers need people to run, test and govern these systems whether or not any one lab thrives. In our view, the skills behind these titles (shipping software, writing evaluations, owning workflows, managing risk) are likely to outlast the titles themselves.

How to position yourself for a fast-growing AI role

Pick a role, then prove it. A published eval set, a working agent or a short governance memo says more than a certificate.
Use your domain. New roles tend to reward people who know the field AI is entering. A nurse who builds evals for clinical summaries stands out.
Watch where postings appear first. AI companies often post on their own careers pages before large boards. On the 1752vc careers board, the AI jobs track lists roles at AI companies with each employer's own posting date, so filtering to Past 7 days shows what is new.
Search new titles by keyword. Try "forward deployed", "evals", "AI operations" and "AI governance" alongside the usual titles, then filter the AI track by level to find roles you can realistically land now.

Our pillar guide on how to get a job in AI has a full 90-day plan. For the broader picture beyond AI, see our sibling guide on the fastest-growing tech careers.

Common mistakes

Chasing the hottest title. Fast growth from a tiny base may mean few openings near you.
Assuming "AI job" means research. The fastest-growing doors are in engineering, deployment, operations and governance.
Ignoring where AI substitutes. If AI can already do most of a role's tasks, growth there may be fragile, especially at entry level.

Where we land

AI is creating real jobs, quickly, and most of them are about making AI work inside companies rather than inventing it. The new titles are small and fast-moving, which suits people willing to learn in public.

It's our view, not a forecast to bank on. Some of these titles will likely be renamed by 2030.

The bottom line

The safest AI career is not the one with the hottest title. It's the one where AI makes you more useful, not less.

Titles come and go.

The work of making AI actually work is just getting started.

Key takeaways

AI engineer topped LinkedIn's Jobs on the Rise 2026 US list, and the BLS projects data scientists (+34.6 percent) and research scientists (+21.8 percent) among the ten fastest-growing occupations from 2025 to 2035.
Sources disagree on the size of AI demand because they measure different things, from 2.56 percent of US postings (Lightcast, 2025) to 6.74 percent (Indeed, August 2026).
New roles such as forward-deployed engineer, evals engineer, AI trainer, agent operations and AI governance are growing from a small base.
WEF expects 170 million jobs created and 92 million displaced globally from 2025 to 2030, but early-career workers in AI-exposed jobs face a tighter market.
In our view, the strongest bets are roles where AI complements your work and where you can show proof without a PhD.

Frequently asked questions

What jobs will AI create in the next five years?

Employers surveyed for the WEF's Future of Jobs Report 2025 expect growth in AI and machine learning specialists, big data specialists and software developers through 2030. Newer titles include forward-deployed engineers, evals engineers, AI trainers, agent operations roles and AI governance specialists. Demand for data center and power infrastructure workers is also expected to rise with AI adoption, per the BLS.

What is the fastest-growing AI job right now?

By LinkedIn's Jobs on the Rise 2026 ranking for the US, AI engineer was the fastest-growing title over the prior three years. In the BLS's 2025 to 2035 projections, data scientists are the fastest-growing AI-related occupation at 34.6 percent, and computer and information research scientists follow at 21.8 percent. Both rank among the ten fastest-growing US occupations overall.

What AI careers are worth starting in 2027?

In our view, roles that complement AI rather than compete with it: AI engineering, forward-deployed or solutions engineering, evaluations, agent operations, AI governance, and data science in a specific domain. Each lets you show proof of skill without a PhD. Treat any list as a starting point, since titles and demand in this field shift quickly.

Will AI eliminate more jobs than it creates?

The WEF's 2025 employer survey expects a global net gain, with 170 million jobs created and 92 million displaced from 2025 to 2030. The BLS expects AI to reduce demand in some office, administrative and media occupations. Research on ADP payroll data finds early-career workers in AI-exposed jobs have lost ground, so the gains and losses may fall on different people.

What does a forward-deployed engineer do?

A forward-deployed engineer works directly with a customer to get a software or AI product running inside the customer's own systems and workflows. The job mixes coding, integration and consulting: scoping the problem, building a working deployment and feeding lessons back to the product team. It suits engineers who prototype fast and like working with clients.

Sources

Disclaimer: This guide is for general education only and is not legal, tax or investment advice. Laws, market data and program terms change, so it may not reflect the latest developments or fit your situation. Treat it as a starting point, not a source of truth, and talk to a qualified lawyer, accountant or financial adviser before you make decisions.