Eighty-three percent. That is the share of American adults who now expect AI to eliminate more jobs than it creates, according to a new Elon University survey. If you have felt that worry yourself, you are in very large company. So, is AI taking jobs? Last week the Fed answered with something more useful than a yes or no: it showed how it is happening. Not mostly through dramatic layoffs, but through openings that never get posted. Here is what the numbers say, both sides of the debate, and a 30-minute career check you can do this week.

What happened

Is AI taking jobs? What the Fed just found

Eight times a year, the Fed’s 12 regional banks interview business owners and managers and publish what they hear in the Beige Book. The October 14, 2026 edition, based on information collected through October 5, said most districts saw little or no net change in overall employment. Then came the line that made headlines: five districts reported firms that had trimmed staff through attrition, delayed hiring, or expected to create fewer jobs after AI made their teams more productive.

The district reports show what that looks like on the ground:

  • Chicago: a business services firm said AI productivity gains let it shrink headcount through attrition, meaning people who left were not replaced.
  • Cleveland: two banking contacts said automation and AI reduced staffing needs by making certain roles more productive.
  • New York: some larger organizations were cutting jobs or pausing hiring as they built AI into their operations.
  • Minneapolis: IT staffing contacts said AI was not causing layoffs, but was slowing the creation of new jobs.

As 24/7 Wall St. (October 15, 2026) pointed out, the Beige Book does not say AI caused widespread layoffs. It describes a slower, quieter shift: the job that opens up when someone retires or moves on, and simply stays closed.

The numbers

How many jobs has AI cut in 2026?

The outplacement firm Challenger, Gray & Christmas tracks the reasons companies give when they announce job cuts. Its September 2026 report, released October 1, counted 573,195 announced U.S. job cuts from January through September. AI was cited for 120,136 of them, about 21%, making it the leading reason so far this year.

The monthly picture is calmer. In September, AI was cited for 3,961 cuts, about 9% of the month’s total, and ranked fifth among reasons. Employers also announced plans to hire 90,787 people that month.

Myth: AI is replacing workers overnight. Fact: announced AI-linked cuts are real and growing this year, but the broader job market has barely moved. The clearest signal is fewer new openings, which is harder to see and matters most to people looking for their first or next role.

Both sides

Is AI really to blame for job losses?

The case that AI is reshaping work now. Five of 12 Fed districts heard it directly from employers. AI is the top stated reason for announced cuts in 2026. And 24/7 Wall St. reports that Amazon and Meta are pursuing job reductions while investing heavily in AI. When companies get more done with the same team, the first thing to slow is hiring, and that pressure lands on new graduates and career changers.

The case that AI is getting too much of the blame. Most districts reported little change in overall employment. In the same Challenger report, the firm’s own comments pointed to high energy costs, a possible rate hike, rising healthcare costs and the war in Iran as pressures on employers. And as 24/7 Wall St. notes, layoffs are not always about AI: companies also cut because of weak demand, financing costs and past overhiring. Meanwhile, the Beige Book says shortages persisted for specialized, technical and skilled trade workers, and new kinds of work are appearing: per 24/7 Wall St., a Dallas staffing firm used AI agents to run interviews that its own employees then reviewed.

Where the evidence is still unclear: nobody can yet separate how many unfilled roles are AI and how many are a cautious economy. Both things can be true at once, and the honest answer is that it depends heavily on the kind of work you do.

What people feel

What workers say they need

The Elon University survey (released October 13, 2026; 1,158 U.S. adults, fielded August 20 to 24 by SSRS) found more than fear. Among workers, 66% are either comfortable with AI at work now or would be with more training. Half are highly confident they can adapt their skills, though only 40% feel as sure about steering their own career path.

Their top requests are practical:

  • Training on AI’s limits and when human judgment should lead (52%).
  • Clear, updated AI use policies (49%).
  • Advice on which roles will grow or shrink (45%).
  • Support for skills that complement AI (45%).

And it is already part of daily work. The U.S. Census Bureau found 56% of workers used AI for at least one task, and about a third of last-week users said it saved them one to two hours. Those hours are the real story: who decides what they are used for?

Your takeaway

What to do if you’re worried AI will take your job

Information shrinks the fear. Set aside 30 minutes this week:

  1. List your week. Write the 10 tasks that take most of your time.
  2. Sort them. Mark each as routine (drafting, summarizing, data entry), judgment (decisions, quality checks) or human (clients, coworkers, care).
  3. Practice on one routine task. Do it with AI this week and check every result yourself.
  4. Grow the other two columns. Decisions, quality checks and relationships are the parts of work AI does not do for you.
  5. Ask for what workers asked for. Request training and a written AI policy from your manager. You are not alone in wanting them.
Try this prompt 
I want an honest look at how AI could change my job.
My role: [job title and industry]
My 10 most time-consuming weekly tasks: [list them]

1. Sort each task into: routine (AI could speed it up),
   judgment (needs my decisions), or human (relationships).
2. For two routine tasks, show me how to do them with AI
   and what I must double-check myself.
3. Suggest three skills that make my judgment and human
   work more valuable, with one free way to practice each.
Do not guess about my company. Ask me if you need context.

If you lead a team, this is the moment to bring people in rather than leave them guessing. Our AI training for teams covers exactly what workers asked for: limits, policy and hands-on practice. Running a small business? Read will AI replace small business jobs and whether AI is worth it for small business next.

Free resources

Free resources for understanding AI and your job

FAQ

Frequently asked questions

Is AI taking jobs right now? +

Some, mostly quietly. The Fed’s October 14, 2026 Beige Book says firms in five of 12 districts reduced staff through attrition, delayed hiring or expected fewer new jobs after AI raised productivity. Most districts still reported little or no net change in employment.

How many layoffs have been blamed on AI in 2026? +

Challenger, Gray & Christmas counted 120,136 announced U.S. job cuts citing AI from January through September 2026, about 21% of all announced cuts. In September alone, AI was cited for 3,961 cuts and ranked fifth among reasons.

Which jobs is AI affecting first? +

Beige Book contacts pointed to roles where AI raises output per person, such as some banking, business services and IT staffing work, plus cleaning robots in janitorial work. Shortages continued for specialized, technical and skilled trade workers.

Do most Americans think AI will take more jobs than it creates? +

Yes. In an Elon University survey fielded August 20 to 24, 2026, 83% of U.S. adults said they expect AI to eliminate more jobs than it creates. Two in three workers said they are comfortable with AI at work or would be with more training.

How can I protect my job from AI? +

List your weekly tasks, find the ones AI speeds up, and practice doing them well with AI while keeping the judgment and relationship parts of your role. Ask your employer for training and a clear AI policy, which are workers’ top two requests in the Elon survey.

Your team wants training and a clear plan. Give it to them. Pixel Pinkscape runs hands-on AI workshops that show people where AI helps, where human judgment leads, and how to write a simple AI policy together. Every project starts with a free 15-minute fit call. Bring AI training to your team.

Sources: Beige Book (October 14, 2026) · Challenger, Gray & Christmas (October 1, 2026) · Elon University (October 13, 2026) · 24/7 Wall St. (October 15, 2026) · U.S. Census Bureau (August 11, 2026, revised September 24, 2026)

Join the conversation

What do you think?

Have you seen AI change hiring where you work, through fewer openings, new tasks or new roles? Tell us what it actually looks like from your desk.

Comments are reviewed before they appear. Keep it kind, and please don’t share private information or links.

Keep the conversation going and follow Pixel Pinkscape: