How AI Will Affect the US Job Market and Your Career

Start with tasks, not jobs

When people ask "will AI take my job," they usually picture a whole job vanishing. That is not how it mostly works. Every job is a bundle of tasks. AI takes over some tasks, speeds up others and leaves the rest. So the better question is, which parts of your work can a machine do, and what is left for you?

The McKinsey Global Institute report Generative AI and the future of work in America put a number on this. It estimated that activities accounting for up to 30 percent of hours currently worked across the US economy could be automated by 2030, a trend that generative AI accelerates. It also projected about 12 million additional occupational transitions in the US by 2030, meaning people moving from one type of job to another.

Which kinds of work are most exposed

AI is strongest at work that is repetitive, rule based and text or data heavy. In the same McKinsey report, declines in food services, customer service and sales, office support and production work made up almost ten million of those 12 million expected transitions. Office support is the one most people in our industry should watch.

• Routine admin: data entry, scheduling, basic document handling and record keeping.
• Scripted customer contact: first line support, order status, simple claims questions.
• First drafts: standard reports, simple code, boilerplate documentation and summaries.

The US Bureau of Labor Statistics analysis of AI, IT and employment for 2024 to 34 shows the same direction. It projects customer service representatives to decline 5.5 percent, and claims adjusters, examiners and investigators to decline 5.1 percent, with AI productivity gains named as one reason demand will be limited for some office roles.

Which jobs are growing

The same BLS analysis projects total US employment to grow 3.1 percent over that period. Compare that with the technology roles:

• Data scientists: projected growth of 33.5 percent.
• Information security analysts: projected growth of 28.5 percent.
• Software developers: projected growth of 15.8 percent, the largest gain in number of jobs on the list.
• Medical records specialists: projected growth of 7.1 percent, which matters for healthcare IT and revenue cycle teams.

So the picture is not "AI destroys jobs." It is more like AI moves demand. Somebody has to build, secure, govern and run these systems, and somebody has to check what they produce. McKinsey's later research, Agents, robots, and us, found that demand for AI fluency, meaning the ability to use and manage AI tools, jumped nearly sevenfold in two years in US job postings.

The entry level squeeze

Here is the part that worries me for young professionals. The tasks AI handles best are often the tasks we used to give juniors. Writing first drafts of test cases. Cleaning data. Basic code fixes. Summarizing meeting notes. Those were how people learned the job.

Research from the Stanford Digital Economy Lab, Canaries in the Coal Mine, used payroll data and found a relative decline in employment for workers aged 22 to 25 in the most AI exposed occupations, while more experienced workers in the same occupations held steady or kept growing. So experience is becoming more valuable, and the first rung of the ladder is harder to reach.

If you are early in your career, this does not mean give up. It means you need to show you can do more than the tasks AI already does. Real projects, contract work and a clear story about the problems you solved matter more than ever.

How to reskill in a practical way

Reskilling sounds big and expensive. It does not have to be. Here is a simple plan that works whether you are a developer, BA, QA analyst or project manager.

1. Map your own tasks. Write down what you do in a normal week. Mark which tasks AI could do today. Those are the ones to hand off, not to defend.
2. Get fluent with the tools in your field. Coding assistants for developers, AI test generation for QA, AI drafting for BAs and PMs. Use them daily on real work.
3. Move toward judgment work. Requirements, design decisions, stakeholder conversations, risk calls and quality review. These are harder to automate.
4. Add a domain. Healthcare, finance, insurance or supply chain knowledge makes you harder to replace than general skills alone.
5. Prove it. A certification, a portfolio project or a contract assignment that shows you applied the new skill. See our guide on building proof of skills.

A word for mid career professionals

If you have ten or fifteen years in IT, your experience is an advantage, as long as you keep it current. The risk for experienced people is not AI itself. It is staying with old tools while the team around you moves on. So put a little time every week into learning, and you keep the advantage.

What companies should do

For employers the lesson is similar. Do not cut all entry level hiring because AI can do junior tasks. If you stop training juniors, you will have no mid level people in a few years. Plus, redesign roles around AI so people spend time on review and decisions, not on copy paste work. And invest in training your current team, its usually cheaper than replacing them.

If you need help staffing new AI, data or security roles, or backfilling while your team reskills, our IT staffing team can help.

The bottom line

AI will change most jobs in the US job market, and it will remove some. But the research points to movement more than collapse. The people who do well will be the ones who learn the tools early, move toward judgment based work and keep proof of what they can do.

Written by

Israr Kazmi

Israr Kazmi is the founder of iCareTechno, the IT staffing, recruitment and consulting division of iCareBilling LLC in Chicago. He writes about hiring, vendor relationships and careers in IT.

More about the author