Ford admitted it spent three years fixing what its AI got wrong. Two dozen others spent Thursday launching a coalition about it.

On Wednesday, Ford Motor Company told reporters something most companies will not say outloud which is that it had bet on AI to do work... and then it had to hire people back to fix the results. Over three years, Ford brought in more than 350 experienced engineers, many of them former employees, to retrain the AI design tools that had been built to replace them and to catch the quality problems those tools were missing. "Mistakenly we thought that by just introducing artificial intelligence and ingesting the design requirements that we had, that that would produce a high-quality product," said Charles Poon, Ford's VP of vehicle hardware engineering. The next day, Ford came in first among mainstream brands in J.D. Power's 2026 Initial Quality Study, its best showing since 2010, past quality stalwarts like Toyota Motor Corporation and Honda. CEO Jim Farley called it an overnight success that was four years in the making.

The same day, a coalition called RAISE US launched to do something adjacent and much larger: prepare American workers for the jobs AI is eliminating. It's a serious effort, nonpartisan, co-chaired by former Commerce Secretary Gina Raimondo, who's also its CEO, and former Indiana Governor Eric Holcomb, with more than $500 million committed toward a $1 billion goal and an advisory board that runs from MIT economist David Autor to AFL-CIO president Liz Shuler to Paul Ryan. The members are the names you'd expect: Amazon, Microsoft, OpenAI, Anthropic, IBM, Cisco, General Motors, Eli Lilly and Company. Raimondo's framing was that, "There's an enormous amount of money and focus right now on winning the technology: the chips, the models. There's not enough attention on securing the future for the American worker."

Here's the tension, and it's the thing every media outlet caught, too: Amazon and Microsoft, two of the anchor members, are running some of the largest AI-attributed layoffs happening right now. So the easy read, the one already all over LinkedIn, is hypocrisy. Companies cut workers, then fund a coalition to look like they care.

I don't think that's the best read, because it lets us off the hook for the harder question.

A coalition like this is one piece of a strategy and what determines which kind, for any given member, is what that company does with its next round of cuts. RAISE US hands every member a true, defensible sentence: we are investing in the transition. But the thing your stakeholders will actually test is whether that sentence is the whole strategy or just the comfortable half of it. Ford's version of "we care about the workforce" cost 350 salaries and a reversal of its own automation bet. The membership version costs a logo on a website and a line in a press release.

Most of us won't be in RAISE US; it's a club of two dozen of the biggest companies in the economy. But we've all got some version of it - a retraining program, an AI taskforce, a "future of work" pledge - that's meant to be your answer when an employee or a reporter asks what you're doing about AI and people's jobs. The problem is that no single program is the answer. People don't judge you on the thing you announce, they judge you on everything you're doing at the same time: the training program and the layoffs and the hiring freeze and the AI tool you just dropped on everyone's desk. Lean on one program to carry that whole story, and the juxtaposition between it and everything else is what people focus on.

A workforce promise with nothing behind it doesn't protect the company but it does hand everyone the line they'll quote back at you after the next layoff. So your job here isn't to wordsmith the announcement, it's to (once again) be the squeaky wheel in the room, asking the unglamorous question again and again: okay, but what are we doing, what's the long-term plan, and how does this program serve it? Narrative can't just be about words but about actions and substance. If the substance is real, find it and point to it the way Ford pointed to a reversed call and a quality score that moved positively. If it isn't there yet, say so before anyone hits publish because the question your people are really asking is "what did you do?" And if the honest answer is nothing, are you going to be the one who says that out loud, or the one who explains it after the layoff says it for you?

Same crack, same directive, two airline responses

On June 22, EASA, Europe's aviation regulator, issued an emergency airworthiness directive ordering urgent inspections of 16 Airbus A380s after cracks were found in the wing mid-spars, the beams that carry the load across the wing. The regulator said the cracks could "reduce the structural integrity of the wing." Fifteen of the affected aircraft belong to Emirates, one to Qantas. Five of the Emirates jets had to be inspected before their next flight; the rest within 25 flight cycles. The directive took effect June 24, and within two days all three companies had said something.

If you run consumer communications, your instinct here is almost a reflex: a beloved passenger brand just got named in a public safety story, so you reassure the flying public. You lean into "our planes are safe," because nothing feels worse for a consumer brand than being the headline on a safety scare.

But here's what Emirates put out. It said it would "comply and carry out the inspections required in accordance with the airworthiness directive," that it was "in close contact with Airbus and the relevant authorities to minimise any disruption to the operating schedule," and that its A380 fleet "remains safe to operate." Now read that as a regular human person with a Dubai flight booked next week. It tells you almost nothing ... which planes, your flight, what happens next, all unanswered. It's compliance language, written to satisfy a regulator and a lawyer, not to settle a nervous flyer. By the consumer playbook, it's a miss.

But go looking for the panicked Emirates passengers and they're not really there - and believe me, I looked. No rush to rebook, no wave of "is my flight safe," no Reddit worry. The travel press was already doing the reassuring on its own, telling readers the worst case is an aircraft swap, not a cancellation. The flying public mostly shrugged.

So who actually moved on this? The market. Emirates grounded five of its highest-capacity aircraft, the ones flying its biggest long-haul routes, Los Angeles, Houston, at a moment when it was already down roughly one in six June flights because of the Gulf disruption. So this isn't really a passenger-reassurance story but a capacity-and-competitiveness story, and the people paying attention are the financial and trade press, the analysts, the partners and lessors watching whether Emirates can fly its schedule.

Read that way, Emirates' "remains safe to operate" is a disciplined one. It gives the flying public the one true thing they needed and gets out of the way, so Emirates doesn't over-invest in calming an audience that wasn't alarmed to begin with and doesn't hand operational detail to the audience that was actually watching. Whether or not anyone in Dubai reasoned it through exactly like that, the shape of the response fits the audience that mattered, not the one the safety framing pulls you toward. (Qantas, by contrast, had nothing to manage; they had one A380, already in a hangar in Dresden, no schedule hit. It could be specific because specificity cost it nothing, but that's not better communication; it's a different problem.)

So the lesson is yes, "know your audience" but everyone says that. It's also a more uncomfortable lesson which is the brand-protective reflex points you at the wrong room. A public safety blemish screams "reassure the consumer," and that pull is loudest exactly when the real stakes are concentrated somewhere narrower - investors, regulators, a handful of big customers. Optimize for the broadest audience and you spend your best energy on the people least likely to act, while underserving the few who'll decide whether this actually costs you. The hardest discipline in a moment like this is to not comfort everyone, but to figure out who moves the needle and aim there.

So before your next "everyone's watching" moment, the question to force in the room isn't "how do we calm the public." It's "who actually moves on this, and are we about to optimize for the biggest crowd instead of the one that matters?"

Your CEO will get asked about Europe. "Macroeconomic headwinds" is not going to cut it

Boston Consulting Group (BCG)'s new June report, built on a survey of more than 20,000 consumers across 11 European countries, found that 53 percent are worried about their daily personal finances, up from 40 percent in 2024. Nearly two-thirds, 63 percent, say they'll only buy at a discount or actively hunt for deals; 62 percent will switch brands to save money. The categories holding up are groceries and household essentials. The steepest pullbacks are in discretionary spend things like fashion, alcohol, packaged snacks. The Wall Street Journal's framing this week was blunter than BCG's: Europeans are "terrified" of spending money.

One caveat is that this is a sentiment survey, and BCG itself notes that actual spending tends to run more moderate than stated intent, especially in a downturn. So "63 percent discount-only" isn't like a fixed fact about behavior. What makes it matter though is that it isn't a one-year blip either. European economic pessimism has now climbed three years running: 49 percent in 2024, 54 percent in 2025, 56 percent in 2026. The trading-down is riding a trend, not a spike, and that's the distinction that should change how your CEO talks about it.

Because the problem isn't the marketing campaign but the earnings call. Companies with real European exposure are about to explain softer Q2 and Q3 numbers, and the house language like "macroeconomic headwinds," "cautious consumer," "softer conditions" does not hold up against data this specific. A consumer who has spent three years trading down and switching brands for price isn't hesitating, she's waiting for confidence to return before resuming old habits. And the difference isn't cosmetic, either: "cautious" implies the revenue comes back when sentiment improves. "Trading down for three years and counting" implies your pricing and positioning are the thing that has to change. Those are two different promises to investors, and you can't make both.

So here's the prep work, and it's more than just swapping in better adjectives.

First, pull the country data, not the European average. BCG breaks this out by market, and the spread is enormous; in last year's read, pessimism ran north of 70 percent in France and Spain and below 40 percent in Scandinavia. "The European consumer" is not a person. Map the behavioral data to the markets where you actually carry revenue, and walk your CEO in with that, not the aggregate.

Second, kill the euphemism and pick a lane. Decide internally, before the call, which story is true for your business: demand that returns when conditions ease, or demand that's structurally repriced. Investors and employees can both smell a hedge so pick the version you can defend with your own numbers and say it plainly, including the uncomfortable one, if that's the true one.

Third, make the internal and external versions match. If you tell investors the consumer is "temporarily cautious" while your own teams are quietly told to plan for sustained trade-down, that gap will come up in a town hall, in a leak, in a Glassdoor review. Consistent framing across audiences isn't a nicety here.

The CCO who walks into earnings prep having mapped the behavior to specific markets, and having forced the "temporary or structural" question before the CEO gets asked it on the record, is the one whose CEO sounds like they're running the business instead of narrating it from a distance. The rest will reach for "headwinds," and then someone on the call will ask the obvious next question. The only real choice is whether you'd rather answer it in the prep room or live.

Keep Reading