Google DeepMind workers in London have voted to unionize in a bid to prevent the company's AI technology from being used by Israel and the US military. In a letter to Google management on Tuesday, the employees called for the recognition of the Communication Workers Union (CWU) and Unite the Union as joint representatives, with 98% of CWU members at DeepMind voting in favor of the move.
See also: Google signs secret AI deal with Pentagon
“We don’t want our AI models to be complicit in violations of international law, but they are already aiding Israel,” an anonymous DeepMind in a statement shared by the CWU. “Even if our work is used only for administrative purposes, as leadership has repeatedly told us, it still helps make genocide cheaper, faster, and more efficient. This must stop immediately, as must the harm to Iranians and human lives everywhere.”
The unionization effort represents an escalation of worker activism that began years before the current vote. In 2018 , thousands of Google employees signed a petition protesting Project Maven , a Pentagon contract that used artificial intelligence to improve drone strikes. This led Google to refuse to renew the contract and to pledge not to develop artificial intelligence for weapons and surveillance.
Google DeepMind's controversial military contracts
The immediate trigger for the unionization was Google ’s $ 1.2 billion Project Nimbus deal with the Israeli Ministry of Defense, announced in partnership with Amazon . Press reports describing how Google ’s cloud services and artificial intelligence technologies were being deployed in Israel ’s military operations fueled the workers’ concerns.
In May 2024 , nearly 200 DeepMind employees had already signed an open letter asking the company to withdraw from military contracts, but management rejected their demands despite meetings. The union movement gained momentum after Google 's decision in February 2025 to reverse its weapons commitment, with employees describing feeling " cheated " by the reversal.
See also: Google employees ask Sundar Pichai to deny use of military AI for secret purposes
If successful, the unionization effort would secure representation for at least 1,000 employees based at Google DeepMind ’s London headquarters . Management now has 10 business days to voluntarily recognize the union efforts before formally initiating legal proceedings to force recognition. The unionization effort includes specific demands that employees want Google to address , which include a clear commitment not to pursue the development of weapons, technologies, or contracts that harm or track people, negotiations around the use of artificial intelligence that “ materially impacts our roles, workloads, or job security ,” and the right for employees to avoid projects that violate their “personal moral or ethical standards .”
DeepMind employees worldwide are also considering protests and “research strikes,” where they avoid working on improvements to Google ’s AI services like the Gemini AI assistant, as part of a broader campaign against Google ’s military-industrial AI contracts . This comes a week after hundreds of Google employees signed an open letter to CEO Sundar Pichai asking the company to refuse to sign classified AI contracts with the Pentagon .
Shortly after, Google – along with OpenAI and Nvidia – signed agreements that reportedly allow the US Department of Defense to use their AI models for “any legitimate government purpose.” In 2024, the company laid off more than 50 employees in response to a protest over Google to the Israeli government.
See also: Google DeepMind opens its first AI campus in Seoul
“ This is a truly important moment where tech workers at Google’s cutting-edge artificial intelligence lab are connecting with some of the most oppressed people in communities around the world in meaningful ways, based on core values of solidarity and unionism ,” said John Chadfield , CWU ’s national officer for tech workers, according to The Verge .
