Google DeepMind workers in UK vote to unionize amid deal with US military | US news

Workers who develop Google’s artificial intelligence products in the United Kingdom have voted to unionize, in part because of concerns about a deal between the company and the U.S. military announced last week.

In a letter expected to be addressed to management on Tuesday and shared exclusively with the Guardian, workers at Google DeepMind, the company’s AI research lab, requested recognition from the Communication Workers Union and Unite the Union as joint representatives of the UK-based lab’s staff.

UK workers at DeepMind voted to unionize in April. One of the workers said he was particularly motivated by reports that Google was close to a deal with the Defense Department and pointed to America’s “capricious war in Iran” and the Trump administration’s feud with Anthropic as indications that the department is “not a responsible partner.” The deal was finally announced on Friday.

“I joined the union because of my concerns about the use of AI to reinforce authoritarianism, whether through military applications or surveillance, both abroad and domestically,” added the worker, who requested anonymity for fear of reprisals. “By unionizing, we are taking the traditional route of allowing workers to organize and have a say. »

Another worker, who also requested anonymity, said many of the company’s employees struggled to come to terms with what they saw as their complicity in Israel’s war in Gaza. The company provided the Israeli military with increased access to its AI tools from the early days of the Gaza war, the Washington Post reported last year, and in 2021 it signed, along with Amazon, a $1.2 billion cloud computing contract with the Israeli government.

“Our technology has helped the Israeli army,” said the second British employee, referring to the Israeli army. “I want AI to benefit humanity, not facilitate genocide. »

Google did not respond to a request for comment.

Concerns among Google employees and investors have been growing for years, but were particularly heightened after the company backed away from its pledge last year not to develop weaponized AI. The development was a major motivator for the Google DeepMind workers’ union in the UK, two of them said. While small groups of Google employees have already unionized in the United States, British workers are the first at a “border” AI lab to seek union recognition, they said. Google DeepMind is headquartered in London but has around a dozen offices across North America and Europe. At least 1,000 workers will be represented if the company recognizes the union, according to union officials.

On Friday, the Pentagon confirmed that it had reached agreements with seven major AI companies, including Google. Others included SpaceX, OpenAI, Nvidia, Reflection, Microsoft and Amazon Web Services. Anthropic, whose technology is widely used by the US military but which has fallen out with the Pentagon over future contracts, was notably absent from the group.

“These agreements accelerate the transformation toward establishing the U.S. military as an AI-driven fighting force and will strengthen our warfighters’ ability to maintain decision superiority in all domains of warfighting,” Department of Defense officials said in a statement.

The Trump administration has pushed AI companies to make their tools available on classified networks without the standard restrictions they apply to users. Google’s contract with the Pentagon reportedly includes language stating: “The parties agree that the AI ​​system is not intended and should not be used for domestic mass surveillance or autonomous weapons (including target selection) without appropriate human oversight and control.” » But this language is not binding, and the agreement also states that Google does not have the right to control or veto the government’s “lawful” operational decisions.

Workers who voted to join the union said they did so to pressure Google to meet demands already made by other employees at the company, including that it agree not to develop technology “whose primary purpose is to cause harm or injury to people,” establish an independent ethics watchdog, and grant workers the individual right to refuse to contribute to projects on moral grounds. If the company refuses, they said, it would consider protests and “research strikes,” in which staff refrain from work expected to significantly improve core products such as Gemini, Google’s AI robot, while avoiding detection by continuing to make smaller updates.

Google workers are increasingly vocal about their opposition to weaponized applications of their technology. Last week, while the deal was in progress, more than 600 Google employees signed an open letter to CEO Sundar Pichai demanding that the company not make its AI systems available for classified use.

“We want AI to benefit humanity; and not that it is used in an inhumane or extremely harmful way,” they wrote. “Making the wrong choice now would cause irreparable damage to Google’s reputation, business and role in the world.”

Tech workers are increasingly questioning management over the use of technology they helped develop. In 2024, Google fired 50 workers who protested Project Nimbus, the 2021 contract with the Israeli government. At Microsoft, which the Guardian says has supplied Israel with cloud storage used for mass surveillance of Palestinians, workers occupied the company’s campus with signs reading “No Work for Genocide.” (The company ended the Israeli military’s access to the technology after the Guardian article.)

Investors have also expressed their concerns. A coalition of shareholders holding about $2.2 billion of Alphabet stock wrote a letter to Google’s parent company last week demanding a meeting and greater transparency about Google Cloud and AI deployments in “high-risk” settings. They cited concerns about the company’s provision of services to U.S. immigration authorities, as well as Project Nimbus, and raised questions about “the effectiveness of policy guardrails, internal escalation processes, and board oversight of AI deployments in conflict-affected or security-sensitive environments.”

In 2018, Google also faced widespread employee protests over a military contract known as Project Maven, in which the company agreed to develop AI products for the Pentagon’s analysis of drone images. In response to this backlash, the company did not renew the contract in 2019 and released a set of principles for its AI work, which included a now-abandoned commitment not to design AI for weapons. Palantir took over the Maven project, which continues today.