I've Watched Ford Drift Twice — Six Years Apart

Ford is rehiring the engineers it just let go. In July 2026, the automaker started bringing back hundreds of experienced human engineers to fix quality problems its automated systems couldn't catch. Ford's own VP of vehicle hardware engineering put it plainly: "Artificial intelligence is a fantastic tool, but it's only as good as the information you use to train it."

Reading that felt familiar. Because this is not the first time I have watched Ford drift.

Drift Is Not a Crisis. It's a Direction.

Product drift is what happens when a team slowly abandons the thing that actually works in favor of the thing that's currently exciting. It rarely looks like a mistake in the moment. It looks like progress. It looks like moving fast, cutting cost, chasing the trend everyone in the room is nodding along to.

The tell is always the same: you have good information, and you decide not to use it, because the trend feels more compelling than the truth in front of you.

I saw Ford do this in 2020. I watched them do it again in 2026. Same pattern. Different decade.

2020: The Data Said Stay Home

In the early pandemic, I was on an IBM team helping clients reopen safely. Ford was one of those clients.

We were pulling Watson insights from a lot of messy, real-world data, and one signal jumped out. When we brought in unemployment data, we could see a massive spike in daycare workers losing their jobs — a genuine hockey-stick curve. That number told a quiet but critical story: if daycare workers in an area were unemployed, then factory workers in that same area had no childcare. And workers with no childcare cannot come back to the line, no matter what the reopening plan says.

Daycare wasn't the only reason people couldn't return. But it was a real, measurable one, and it was hiding in plain sight in the data.

Ford didn't act on it.

On day two after the Chicago assembly plant reopened, someone tested positive for COVID-19, and workers walked out. Production halted. The reopening optics that mattered so much collided with the reality the data had already flagged.

This is what makes drift so frustrating to watch. The insight existed. The warning was on the table. The trend — reopen fast, reopen visibly — simply mattered more in that moment than the information did.

At IBM, we took the childcare signal seriously enough to join the lobbying effort for "bubble quarantines" in Michigan. The idea was simple and human: let small, trusted groups of factory workers watch each other's kids so they could safely return to work. It wasn't glamorous. It didn't trend. But it addressed the actual problem the data had surfaced, instead of pretending the problem wasn't there.

2026: The Trend Said Fire the Experts

Fast forward six years. The exciting trend is no longer "reopen at all costs." It's "AI can do it."

Ford laid off hundreds of experienced engineers, citing AI. The logic is seductive and everywhere right now: automate the work, cut the headcount, ride the wave, tell investors you're lean and modern.

Then the quality problems showed up. The automated systems couldn't handle the issues that seasoned engineers catch on instinct — the judgment calls, the edge cases, the "that doesn't feel right" moments that come from years on the job. So Ford reversed course and started rehiring the very people it had pushed out.

Ford isn't alone here. Commonwealth Bank of Australia cut customer service staff for an AI voice bot, watched call volumes climb when the bot couldn't cope, and reversed the cuts. IBM replaced HR functions with AI that handled 94% of routine requests but choked on the 6% involving real ethical judgment — then announced plans to triple entry-level hiring. One survey found 55% of leaders who made AI-driven layoffs admitted those decisions were wrong.

But Ford's version stings more when you've seen it before. Because it's the same drift. The company had something that worked — deep human expertise — and traded it away to chase a trend. The trend promised savings. Reality sent an invoice.

The Pattern Underneath Both Stories

Strip away the specifics and 2020 and 2026 are the same failure:

• In 2020, the working thing was real data. The trend was fast reopening. Ford chose the trend.

• In 2026, the working thing was human expertise. The trend was AI cost-cutting. Ford chose the trend.

Both times, the thing that actually worked was quieter than the thing that was fashionable. Data doesn't hold a press conference. Experienced engineers don't spike a stock price. So they get discounted — right up until the moment reality forces a reversal that costs far more than staying the course ever would have.

That's the real price of drift. It's not just the wrong turn. It's the expensive U-turn later, plus the trust you burned with the workers who walked out, or the engineers you let go and then called back.

How to Catch Yourself Before You Drift

You can't outlaw trends, and you shouldn't. AI is genuinely useful. Fast decisions are sometimes necessary. The problem is never the new tool — it's abandoning what works to grab it. A few questions keep you honest:

• What does our best data actually say? Not the data that flatters the plan. The daycare signal was right there. Are you looking at yours, or looking past it?

• What are we giving up to chase this? Name it out loud. "We are trading experienced engineers for automated systems." Said plainly, some trades expose themselves.

• Is this the trend talking, or the evidence? If your strongest argument is "everyone's doing it" rather than "our numbers support it," that's drift wearing a strategy deck.

• What's the cost of the reversal? Assume the trend might not deliver. Rehiring, walkouts, lost trust — price that in before you commit, not after.

Ford didn't fail because it tried new things. It failed, twice, because it let something shiny talk it out of something that worked. The data was solid in 2020. The engineers were skilled in 2026. Ford had what it needed both times and set it down to chase the trend.

Better product patterns. Fewer regrets later. The regret isn't in the new tool. It's in walking away from what already works to reach for it — and paying full price to walk it back.

Sources: CNBC, "Employers who laid off workers for AI are reversing their decisions," July 1, 2026. CBS News Chicago, "Thousands forced to leave Ford plant on day after it reopened," 2020.