DeepMind UK Staff Seek Union Recognition Over Military AI Ties

LONDON — Employees at Google DeepMind have voted to unionize in a direct challenge to the company's accelerating partnership with the Pentagon and its documented role supplying artificial intelligence tools to the Israeli military during its war in Gaza. The move, which could make DeepMind the first major frontier AI laboratory to unionize, reflects growing worker resistance to the weaponization of technology they helped create.

In a letter sent to management on Tuesday and obtained by multiple outlets, DeepMind staff in the United Kingdom requested formal recognition of the Communication Workers Union and Unite the Union as joint representatives. The vote in April delivered overwhelming support, with 98 percent of Communication Workers Union members at the lab backing the effort. If recognized, the unions would represent at least 1,000 workers at DeepMind's London headquarters, the heart of Google's most sensitive AI research.

The union drive gained urgency after Google confirmed a deal last week with the US Department of Defense for classified military projects. Workers expressed alarm that the agreement could pave the way for autonomous weapons and expanded domestic surveillance, particularly under an unpredictable Trump administration that has shown hostility toward perceived rivals like Anthropic and pursued aggressive foreign policy including what one employee described as a "capricious Iran war." Several staff members cited fears that their research would empower authoritarianism both abroad and at home.

"I have joined the union due to concerns about AI being used to empower authoritarianism, whether through military or surveillance applications, both foreign and domestic," one DeepMind worker told The Guardian, requesting anonymity over concerns of retaliation. "By unionizing, we are taking the traditional route for workers to organize and have a say."

The Pentagon contract is not the only source of unease. Multiple employees said their decision was heavily influenced by Google's documented assistance to the Israeli military since the early days of the war in Gaza. A Washington Post investigation last year revealed that Google granted the Israel Defense Forces expanded access to its AI tools shortly after October 2023. That cooperation built on a $1.2 billion cloud computing contract Google and Amazon signed with the Israeli government in 2021.

"Our technology helped the IDF," a second UK-based DeepMind worker said. "I want AI to be used for good, not to make genocide cheaper, faster, and more efficient."

In a statement shared by the Communication Workers Union, another employee was even more direct: "We don’t want our AI models complicit in violations of international law, but they already are aiding Israel’s genocide of Palestinians. Even if our work is only used for administrative purposes, as leadership has repeatedly told us, it is still helping make genocide cheaper, faster, and more efficient. That must end immediately, as must harm to Iranians and human lives anywhere."

The union letter outlines specific demands that go beyond traditional workplace issues. Workers are seeking a binding commitment from Google to cease development of weapons technologies or contracts that enable harm or mass surveillance. They also want negotiations over how AI tools will reshape their own jobs and workloads, and the explicit right to refuse participation in projects that conflict with their moral or ethical standards. Some DeepMind staff globally are already discussing potential research strikes and in-person protests targeting improvements to products like Gemini.

Google DeepMind CEO Demis Hassabis has not publicly responded to the unionization effort. The company now has 10 working days to voluntarily recognize the unions before organizers pursue legal routes to compel recognition. A successful union would mark a significant milestone in the tech sector, where frontier AI labs have largely operated without collective worker representation despite wielding enormous influence over the future of warfare, surveillance, and public policy.

The developments expose deeper tensions inside Google, which has faced internal dissent for years over military contracts. Many employees say they were already struggling with moral injury from the company's role in Gaza when news of the Pentagon deal broke. For these workers, unionization represents more than better pay or conditions. It is an attempt to reclaim some democratic control over technology that, once developed, can be repurposed in ways that contradict the humanitarian values many researchers believed they were advancing.

This is not an isolated incident. Across the technology industry, AI practitioners are increasingly questioning the unchecked power of a handful of corporations to shape global conflict through opaque deals with governments. The DeepMind union drive stands out because it targets one of the most prestigious and secretive AI laboratories at the precise moment its tools are being integrated into military infrastructure.

Whether Google will accede to the workers' demands remains uncertain. What is clear is that a growing number of the people actually building these transformative systems are no longer willing to remain silent while their labor is deployed in service of endless war and mass surveillance. By organizing through traditional unions, DeepMind staff are asserting that even in the rarefied world of artificial intelligence, workers retain the fundamental right to refuse complicity in atrocities.