Tesla Grünheide was meant to be the showcase of European car making. Instead, in January 2026, Germany’s leading business paper reported that the plant quietly lost around 1,700 workers in two years. Management had said far fewer jobs would go.
So who was right? This story has three voices: what management said, what the union said, and what the numbers showed. Together, they reveal a gap every leader should worry about.
Voice one: what Tesla Grünheide management said
In April 2024, Tesla announced global job cuts of more than 10%. For Grünheide, the official line was that about 400 positions would go. Management also said there would be no mass layoffs.
Then, in March 2025, plant manager André Thierig said there were no plans for production stoppages, job cuts or short-time work at the plant (electrive, citing Tagesspiegel and Handelsblatt, January 2026).
Voice two: what the union said
Back in 2024, the IG Metall union painted a different picture. It alleged that staff faced extremely high workloads and a “culture of fear”. It also said this pressure was driving stress and sick leave (Deccan Herald, citing The Guardian).
At the time, sickness absence at the plant reached up to 17% on some days. Managers even visited some workers at home while they were on sick leave. The plant manager said such visits are common in the industry.
Voice three: what the numbers showed
In January 2026, Handelsblatt reported that Tesla Grünheide employed 10,703 people. Two years earlier, at the works council election, it had 12,415. That’s about 1,700 fewer workers, a drop of 13.8% (electrive, January 2026).
In other words, the reduction went well beyond the roughly 400 roles Tesla had signalled. According to the reports, staff never received an official announcement of cuts on that scale.
Tesla’s response
Tesla disputes the idea of large-scale job cuts. A spokesperson said there had been no significant reduction in the core workforce compared with 2024, and that none is planned. Tesla also said changes in the total workforce, including temporary workers, are completely normal after the factory’s fast ramp-up. It added that production remains stable and that a battery expansion by 2027 should create new jobs.
The statistic behind the gap
In September 2026, The Predictive Index reported on its 2026 AI and People Leadership Survey of 399 managers and 208 CEOs. It found that 74% of CEOs believe their managers can handle hard conversations on their own. Yet only 42% of managers say they know exactly what to say.
That’s the heart of the Grünheide story. When leaders assume the message is landing, and managers don’t know how to deliver it, people fill the silence themselves. Usually, they assume the worst.
What MoodyBot could have shown at Tesla Grünheide
MoodyBot, the AI layer inside Moody At Work, reads anonymous daily mood check-ins across teams. It doesn’t set headcount or manage sick leave. Instead, it tracks how people feel at work every day, with no names attached.
In a plant like this one, MoodyBot could have flagged rising stress on specific shifts or lines. It could also have shown check-in reasons clustering around workload, job security or not being told the truth. Finally, it could have revealed a gap between how permanent staff and temporary staff felt, which the sickness data already hinted at.
Three voices told three different stories. By contrast, anonymous daily data gives leaders one shared picture, early enough to act on it.
| Risk | Publicly reported figure | Model assumptions | What MoodyBot could have saved |
|---|---|---|---|
| Peak sickness absence (2024) | Up to 17% on some days, on a workforce of about 12,400 | Roughly 2,100 people absent on peak days; 10% to 20% of peak absence avoidable through earlier action on workload and stress signals | About 210 to 420 fewer absences on peak days |
| Workforce reduction | About 1,700 fewer workers over two years, versus roughly 400 roles signalled | Tesla says most changes involved temporary staff; costs not disclosed | Not quantified |
| Trust and communication | Reports that staff received no official announcement on the scale of cuts | Not measurable from public data | Not quantified |
| Total quantified | About 210 to 420 fewer absences on peak days |
How we built the numbers: 17% of about 12,400 workers gives roughly 2,100 people absent on peak days. 10% of that gives about 210, and 20% gives about 420. Tesla doesn’t publish plant-level costs, so we measured capacity instead of inventing a money figure.
What every leader can learn from Tesla Grünheide
People don’t need perfect news. They need honest news, delivered by managers who know how to say it. When official messages and lived experience drift apart, trust goes first and absence often follows.
Hidden strain showed up in a very different workplace in the Hannaford overtime settlement.
Tesla Grünheide FAQs
How many jobs did Tesla Grünheide lose? According to Handelsblatt, the plant employed 10,703 people in January 2026, down from 12,415 two years earlier, a fall of about 1,700. Tesla says there was no significant reduction in its core workforce.
Why did the union criticise Tesla Grünheide? In 2024, IG Metall alleged high workloads and a culture of fear that it linked to stress and sick leave. Tesla has rejected claims about health and safety at the plant.
Can managers handle hard conversations? According to The Predictive Index’s 2026 survey, 74% of CEOs think so, but only 42% of managers say they know exactly what to say.
Sources: The Predictive Index, 2026 AI and People Leadership Survey (September 2026); electrive (22 January 2026, citing Handelsblatt, Tagesspiegel and Der Spiegel); Deccan Herald (citing The Guardian); EVwire.
All factual claims are sourced from publicly available reporting. Tesla’s representatives have stated that there has been no significant reduction in the core workforce compared with 2024, that none is planned, and that workforce fluctuations including temporary workers are normal following the factory’s ramp-up. MoodyBot projections represent an independent illustrative probability model based on publicly reported information and are not guaranteed outcomes.