Google Employees Petition Against Taking US Military AI Contracts

A new wave of employee protests has erupted within Google. Hundreds of artificial intelligence researchers have co-signed an open letter, demanding CEO Sundar Pichai to refuse the use of the company's AI systems for secretive missions of the US Department of Defense.

According to Bloomberg, the organizers claim to have collected over 580 signatures, including more than 20 executives at the level of directors, senior directors, and vice presidents, as well as several senior employees from Google's AI research institution, DeepMind. The organizers stated that the letter would be delivered to Pichai this Monday.

The open letter explicitly demands that Google refuse all secretive workloads, on the grounds that once AI tools are deployed in systems that are physically isolated from the public internet, the company will be unable to monitor their actual use.

This letter comes at a time when negotiations between Google and the Pentagon continue to advance. Reports indicate that discussions are ongoing regarding the use of Google AI tools for "all legal purposes." Critics argue that, in practice, this could encompass fully autonomous weapon systems and large-scale domestic surveillance.

This move by the employees adds further uncertainty to Google's future expansion in defense business and the boundaries of compliance.

Employee Concerns: Vague Contract Boundaries and High Risk of Loss of Control over Secretive Systems

The core demand of the open letter is that employees believe Google has failed to draw specific and enforceable red lines regarding the use of AI on secretive networks.

Sofia Liguori, an AI research engineer at Google DeepMind's UK branch, stated that she signed the letter because Google failed to discuss with employees the use of AI in secretive domains. She pointed out that once AI tools are deployed in secretive systems, the company will be unable to track and limit their actual use from a technical standpoint.

"The company's response has always been to trust the leadership to sign good contracts," Liguori said, "but these expressions are very broad." She specifically named the risk of "Agentic AI":

"It's worrisome to see the level of autonomy it can achieve. It's like handing over an extremely powerful tool while also relinquishing any control over its use."

The open letter states: "At present, the only way to ensure that Google does not become associated with such harm is to refuse any secretive workloads. Otherwise, such uses may occur without our knowledge and beyond our ability to prevent them."

History Repeats: From Maven to the Current Boundary Dispute

This employee action is not the first of its kind. In 2018, Google employees strongly protested against the company's participation in the Pentagon's "Maven Project," which aimed to use AI to analyze target objects in drone video footage.

Facing employee opposition and a wave of resignations, Google ultimately formulated new AI principles and decided not to renew the Maven contract.

However, Google gradually rebuilt its relationship with the defense industry afterward and deleted the commitment to avoid using technology in potential harmful applications such as weapons from its AI principles last year.

The organizers of this joint letter stated in their declaration: "Maven is not over yet." They expressed that "workers will continue to organize actions around the weaponization of Google AI technology until the company sets clear, enforceable boundaries."

Google and the Pentagon: Accelerating Cooperation, Doubts About Boundaries

Meanwhile, cooperation between Google and the Pentagon has made substantial progress. In March of this year, Google opened its Gemini AI agent to over 3 million employees of the Pentagon for non-classified use; the Gemini chatbot had been opened in December last year.

Deputy Secretary of Defense for Research and Engineering, Emil Michael, told Bloomberg in March that the Pentagon will start with the uses of the non-classified Gemini AI agent, "and then we will move into classified and top-secret levels," revealing that negotiations for the use of Google AI agents in classified clouds are underway.

Negotiations between Google and the Pentagon on the authorization of