Oliver Blume, CEO of Volkswagen Group, brought the biggest restructuring proposal in Volkswagen's 89-year history to his board last Thursday, reportedly up to 100,000 job cuts worldwide, about 15 percent of the workforce, plus the closure of four German plants. The board voted 12 to 7 against him.

Some context on why a CEO can lose a vote like this: German law requires big companies to give employees half the seats on the supervisory board, the board that has to approve major decisions, and it's a system built on the idea that workers should have a formal say in choices that affect their jobs, and it means a plan like this needs at least some labor support to pass. Blume didn't have any; twelve members voted no and at least two people from the shareholder side, the investors' representatives, the people who'd benefit most from cost cuts, voted against him too.

Blume's predecessor, Herbert Diess, clashed with the unions constantly, and it was labor opposition that ultimately ended his run in 2022, the same thing that ended another VW CEO's tenure back in 2006. Blume arrived as the opposite: a consensus-builder by reputation, seen inside and outside the company as a fresh start after Diess had alienated the workforce, and credited since with an ability to balance every one of VW's stakeholders, unions included.

And when the first restructuring crisis hit in late 2024, he ran the play German CEOs are expected to run: he stood in front of 20,000 workers at a plant-wide meeting in Wolfsburg (German employees are entitled to these assemblies, where management takes questions directly) absorbed the boos, negotiated with the union, and landed a deal. That deal cut 35,000 jobs by 2030, and in exchange management committed to no German plant closures and no forced layoffs through the end of the decade.

The new plan allegedly proposes closing four plants... eighteen months after promising not to.

Employees learned the scale of the new plan from media leaks, which then prompted workers to protest at plants across the country on the day of the vote, and Daniela Cavallo, the elected leader of VW's employee representatives, someone who once got along with Blume, called the board's treatment of the workforce "the height of disrespect." After the vote failed, employee representatives sent management more than 80 questions and Cavallo publicly demanded that Blume come explain the plan to employees himself. Blume's answer came Sunday, in a newspaper interview instead of an internal forum.

So why would a CEO who has literally stood in front of 20,000 booing workers now avoid the room? Probably because this time there is no good answer. Facing the workforce means answering the tough question about why "you promised us no closures". The media interview is what avoidance looks like when the honest conversation requires accounting for a broken commitment. Which points at something we say often here: reputation is an integrity problem before it's a communications problem. No channel strategy fixes a reversal the company hasn't yet explained.

On Tuesday, Meta launched a new AI image generator called Muse Image with a feature that let anyone create AI images of another person just by tagging their public Instagram account. If you're an adult with a public account, your photos were included automatically and turning it off meant finding a setting buried in your account menu, and nothing told you if someone had already used your face. The backlash was immediate. Tech sites published how-to guides for opting out, the actors' union SAG-AFTRA told its members to protect their likenesses, Creative Artists Agency, one of Hollywood's biggest talent agencies, blasted the company. By Friday, three days after launch, Meta had deleted the feature and put out a short statement saying it "missed the mark."

The consent problem is the obvious story, and that's true. But here's the part that matters for communications professionals: this exact mistake had already been made, in public, by a competitor. Last October, OpenAI did nearly the same thing with its Sora video tool -- people's faces were fair game unless they opted out -- and the same actors' union tore into them for it, warning it threatened how the entire entertainment industry makes money. Same feature design, same critics, same complaint, all of it a simple Google search away.... but Meta built and launched it anyway. Which means somewhere in the rooms where this launch got approved, either nobody brought up what happened to OpenAI, or somebody did and got waved off. That's the failure in my mind: the meeting. And it's a failure communications leaders are well placed to prevent, because knowing how these stories end is the job.

Now the part nobody's saying: the embarrassing about-face is the best communications work in this whole mess. Meta didn't pause the feature or promise better controls or a review. It just deleted the thing, in 72 hours And that's exactly why it worked. Within hours, SAG-AFTRA (the loudest critic in the room) publicly praised the company. "We commend Meta for its swift decision to remove the Muse Image feature," its spokesperson said. Your critics can only give you credit for a retreat when it's a real one. The move most companies make instead, "we're pausing the feature while we improve the experience" keeps the story alive, keeps the critics circling, and earns credit from no one. When you take the whole thing back, three sentences is enough.

So if you've got an upcoming launch that smells like it has the potential to upset users, it may be worth doing an audit on what others have gone through before. Call it the precedent audit, and look at a few questions 1) Has anyone shipped something like this before? 2) What happened to them in the first three days? 3) And what exactly makes ours different? If the answer to the third question is "nothing," you may have just saved your company a public retreat.

Germany recorded 4,996 corporate insolvencies in the second quarter (companies going bankrupt) according to the Halle Institute for Economic Research (IWH), a leading German economic institute. That's up 9 percent from the previous quarter and the highest second-quarter number since 2005. The failures cut across construction, real estate, trade, hospitality, and services, and touched roughly 45,500 jobs -- more than the 2005 peak did. The institute's early-warning indicators are still running above last year's levels, and its researchers expect the third quarter to be even worse.

Most multinationals are still describing Europe to investors as a "challenging environment" or "softening macro conditions." If your company updated that language recently, or barely operates in Germany, you may be fine. If nobody's touched your European talking points since the spring, they might be understating things. The companies going under are mostly small and mid-sized firms -- the suppliers, sub-contractors, and local service providers that sit underneath big companies' European operations. So their failures don't show up in your headline numbers until something breaks: a delivery gap, a replacement supplier that costs more, a disruption that takes a quarter to explain to investors. That conversation is much easier to have before it happens.

If you have meaningful presence in Germany, you may want to do a 30-minute sweep of the last several times your CEO or CFO publicly described or spoke to the market. Send those exact quotes to your European operations lead asking, "does this match what you're seeing from suppliers and customers on the ground?" If the answer is anything other than yes, I'd suggest you rewrite the Europe paragraph in your earnings prep materials now and add one line that shows the company is watching the same data everyone else is, something like "we're monitoring elevated supplier insolvencies in Germany and already have contingency sourcing in place."