Why employees turned to the state
The application followed meetings between employees, union representatives and the company about voluntary unionisation. People involved in the initiative said the talks left employees with the impression that management was not taking their demands seriously.
John Chadfield, national technology officer at the Communication Workers Union, told WIRED that employees felt their concerns had been ignored and downplayed. He said management did not negotiate with them directly, instead referring the matter to staff in human resources and employee relations, who treated it as a formal administrative process.
Google DeepMind did not respond to WIRED’s request for comment.
Dispute over ethics and military agreements
The initiative began in February 2025, after Alphabet, Google’s parent company, removed a pledge from its ethical principles not to use AI for certain purposes, including weapons development. One Google DeepMind employee, who asked not to be named for fear of backlash, said that for them and their colleagues, the issue is about returning to the ethical principles the company upheld when it hired them, and to the goal of using technology to benefit people.
Those involved in the union campaign are especially opposed to the company’s agreements with the US and Israeli armed forces. If employees succeed in forming a union, they intend to demand that Google withdraw from those contracts. Another employee, who also requested anonymity, said they were troubled by the direction in which the technology is developing.
The company has previously defended its agreements with government organisations. In April, Google spokesperson Jen Crider told The New York Times that the company was proud to be part of a coalition of leading AI labs, technology companies and cloud providers that delivers services and infrastructure for national security. She also noted that Google supports the shared position of the public and private sectors: AI should not be used for domestic mass surveillance or autonomous weapons without appropriate human oversight.
What employees and unions want
On Thursday, employees involved in the campaign sent a letter to all their colleagues in the UK urging them to take part in the vote. They expect it to take place in the coming months. In the letter, employees said they want a union, collective bargaining and a say in how AI is developed and used.
Those involved in the initiative hope their experience will set an example for employees at other major AI labs in the UK, including OpenAI and Anthropic. The second anonymous Google DeepMind employee said that in an industry with such influence over the economy, workers and regulators have few ways to exercise oversight and impose checks and balances. Employees want to show what can be achieved through collective organisation and a shared position on how technology is used, they said.
Dataist’s take: A vote could give employees a formal mechanism for collective bargaining and for making demands about the company’s military agreements. But for now, the unions have only filed an application: they still need to confirm sufficient employee support, and the vote itself is expected in the coming months.
Daily AI news
Every day we pick what actually matters in AI and explain it plainly — no hype, no filler. Subscribe if you want to follow where the industry is going.
Only what matters — every day
Follow on X