The AI Rehiring Wave: Why Ford, IBM, and a Big Bank Are Rethinking AI Job Cuts

Ford brought in about 350 veteran engineers. IBM is tripling its U.S. entry-level hiring for 2026. Commonwealth Bank of Australia reversed 45 layoffs and apologized to the workers it had let go.

None of these were small experiments. They were public corrections at some of the largest employers in their markets. Two independent surveys suggest the pattern is broader than a few anecdotes: some employers are rehiring after AI-linked cuts, and many leaders now regret AI-related redundancy decisions.

A Robert Half survey of roughly 2,000 U.S. hiring managers found that 32% of those who eliminated a role mainly because of AI later rehired for the same or a similar position. In finance, the figure was 44%. So if you sat in a leadership meeting last year and heard that AI could take over the work and the headcount could come down, there is now a real-world counterweight worth reading closely.

Here is what these companies learned, and what it means for the calls you own.

What Ford learned about the knowledge it almost lost

Ford leaned heavily on AI and automation in quality work while veteran experience left the system. The problem showed up later: the knowledge those workers carried had not been fully captured before Ford tried to rely on AI-driven processes.

Charles Poon, Ford's VP of Vehicle Hardware Engineering, was blunt about the mistake.

"Mistakenly, we thought that by just introducing artificial intelligence and ingesting the design requirements that we had, that would produce a high-quality product," he said. "Over prior years, we didn't pay as much attention as we should have to the experience of our most knowledgeable engineers who have been with us through many product cycles."

So Ford spent about three years hiring back roughly 350 veteran engineers. They now mentor junior staff and rebuild the machine-learning systems that had underperformed.

The turnaround worked. Ford took first place among mainstream automakers in J.D. Power's Initial Quality Study, a ranking it had not topped in about 16 years. CEO Jim Farley described the quality gains as "hundreds and hundreds of millions of dollars of a tailwind for Ford on cost."

Artificial intelligence is a fantastic tool, but it's only as good as the information you use to train it.

Read that as an executive, not an engineer. The value did not live in the tool. It lived in the people who knew why the process worked, and the AI could not learn what they knew until someone captured it first.

IBM found the ceiling at 94 percent

IBM went further into automation than most. CEO Arvind Krishna told The Wall Street Journal in May 2025 that the company had replaced several hundred HR workers with AI agents. Those agents handled about 94% of routine HR tasks.

Then look at the other 6%.

That slice was the work that still needed human judgment, exception handling, and employee context, and the AI could not carry it. So IBM changed course. At Charter's "Leading with AI" Summit in New York in February 2026, Chief Human Resources Officer Nickle LaMoreaux announced the company would triple its U.S. entry-level hiring for 2026.

This is the same IBM that, back in 2023, paused hiring for roles it believed AI could do. Three years later, it is hiring young people at three times the rate.

The 94% number is the interesting part for anyone building a budget. Automation can clear most of the volume. The remaining fraction is where the hard, high-stakes work sits, and staffing for that fraction is a choice, not a rounding error.

The bank that automated a job it had not scoped

Commonwealth Bank of Australia introduced an AI voice bot and moved to cut 45 customer-service staff. The assumption was that call volumes would drop.

They rose instead.

Remaining staff worked overtime. Managers were pulled in to answer phones. In late August 2025, the bank reversed the decision and apologized to the affected workers. The Finance Sector Union welcomed what it called a "backflip" on the cuts.

CBA had automated a task before it understood the task. The bot handled the calls it was built for, but the real demand did not behave the way the plan assumed. That is a scoping failure, and it is expensive precisely because it looks like a safe, obvious efficiency on the spreadsheet.

The pattern under all three reversals

It is tempting to read these as "AI failed." That is not what happened.

AI failed where the work and the human judgment were not designed in first. Ford lost its institutional knowledge before capturing it. CBA automated a job it had not fully scoped. IBM's agents cleared 94% and broke on the 6% that required a person.

In each case the technology did roughly what it was built to do. The mistake was upstream, in the decision about what to automate and what to keep human.

There is a survey that captures how widespread this thinking has become. An Orgvue survey of business leaders found that 39% had made employees redundant because of AI, and 55% of those leaders later admitted the decision was wrong. That research predates this year's rehiring news, so treat it as background rather than a fresh headline. Still, more than half of the leaders who cut for AI came to regret it. That is a hard number to ignore when you are the one signing off on the plan.

What this means for the decisions you own

You do not need to write a line of code to act on any of this. The calls that determine whether AI helps or hurts your P&L are leadership calls.

Design the work before you automate it. The companies that got burned automated a task without mapping how the work actually flows and where human judgment carries the weight. The order is fixed: understand the process, then decide which parts an agent should handle, then build.

Protect institutional knowledge before you cut it. Ford's most expensive lesson was that experience walked out the door unrecorded. Before any AI-driven staffing change, ask a simple question: if this person leaves, does the AI actually know what they know? If the answer is no, you are not ready.

Staff for the 6 percent. IBM automated most of its HR volume and still needed people for the cases that carried human judgment and higher stakes. Budget for the hard fraction. It is small in count and large in consequence.

Measure capability, not fear. The durable skill on your team is not "avoid being replaced by AI." It is the ability to direct AI, catch the cases it gets wrong, and own the judgment. That is what The 7 Levels of AI Proficiency measures: how well your people can work with these tools, from first awareness through leading the work. A team that scores high there is the team that catches the 6% before it reaches a customer.

The lesson from Ford, IBM, and CBA comes down to one thing: the human directing AI is the part you cannot outsource. Companies that cut that human first are the ones writing apology notes and rehire offers now.

Related reading: Level 5: The Captain (Design Thinker).

Sources

1. Employers who laid off workers for AI are reversing their decisions (CNBC)
2. Ford Hiring 350 Engineers After AI Failed Shows Human Value In AI Era (Forbes)
3. Humans Better Than AI Inspectors (Motor1)
4. Commonwealth Bank chatbot fail leads to rehiring (The Register)
5. IBM triples entry-level hires for 2026 despite AI adoption (Tom's Hardware)
6. 55% of businesses admit wrong decisions in making employees redundant when bringing AI into the workforce (Orgvue)

Harrison Painter

Executive AI Advisor. Founder, LaunchReady.ai and AI Law Tracker.

Harrison is an Indiana AI Advisor who helps business owners and executives get their time back by building AI systems that run the work for them. Nearly 20 years in business and author of You Have Already Been Replaced by AI. Creator of The 7 Levels of AI Proficiency.

Connect on LinkedIn