A senior Google executive has stepped down after publicly criticising the company's decision to provide artificial intelligence technology for classified US defence work, claiming the tech giant has moved away from the values that once defined its culture.
René Mayrhofer, Director of Android Platform Security, announced his resignation in a farewell note to colleagues, saying the decision had become "unavoidable" following Google's agreement to allow the US Department of Defense to use certain AI models for classified purposes.
The departure has reignited questions about how technology companies navigate the growing intersection between artificial intelligence, national security and employee values.
According to reporting by India Today, Mayrhofer said the Google he joined in 2017 was fundamentally different from the company he is leaving nearly a decade later.
"I am quite sad that it had to come to this, and desperately hope Google management re-discovers its moral compass," he wrote in the resignation letter.
Pentagon agreement becomes a flashpoint
At the centre of the dispute is a defence agreement announced by Google earlier this year.
According to India Today, Google entered an arrangement that enables the Pentagon to use the company's AI technology for classified work, including applications that may support military planning and intelligence operations.
Mayrhofer said he could not support that direction.
Describing himself as a pacifist, he wrote that he had long made a personal decision not to contribute to offensive military activities.
The executive said the agreement represented a line he could not cross.
"This unfortunately leaves me with the only choice to resign," he wrote.
His resignation highlights a growing challenge for technology companies as AI systems become increasingly valuable for governments, defence agencies and national security organisations.
A criticism of Google's changing culture
The resignation letter extended well beyond Google's defence work.
Mayrhofer portrayed the decision as part of a broader cultural shift inside the company.
Reflecting on his early years at Google, he described an organisation that encouraged debate, embraced diverse viewpoints and expected employees to bring their own values into the workplace.
He also referenced Google's 2018 AI principles, which stated the company would not pursue:
• Weapons or technologies designed to directly facilitate harm to people
• Surveillance technologies that violate internationally accepted norms
• Technologies that conflict with widely accepted principles of international law and human rights
According to Mayrhofer, recent developments suggest Google has moved away from those commitments.
"Google was a different company nine years ago," he wrote.
He added that major strategic decisions were increasingly being made without broader internal discussion.
"None of this is being debated or communicated within the company. It is just decided by top-level management."
Environmental and governance concerns surface
Mayrhofer's criticism also extended to Google's environmental ambitions.
In the resignation note, he accused company leadership of quietly moving away from earlier carbon-neutrality goals as AI infrastructure demands continue to rise.
"Google management has quietly abandoned its goals to become carbon-neutral because of the AI model energy usage," he wrote.
The comments come as major technology companies continue investing heavily in AI infrastructure, including large-scale data centres and advanced computing resources that require substantial energy consumption.
The rapid expansion of AI has increased scrutiny of the environmental impact associated with training and operating increasingly powerful models.
Employee resistance is not new
The resignation is not the first sign of internal opposition to Google's defence-related AI work.
According to India Today, hundreds of employees reportedly signed letters objecting to classified military AI projects after reports of the Pentagon partnership emerged earlier this year.
The publication also reported that Andreas Kirsch, a researcher at Google DeepMind, publicly criticised the agreement and said he was "ashamed" of the decision.
The latest resignation therefore reflects a broader debate that has existed inside Google for years.
The company previously faced employee backlash over military-related technology projects, including concerns around the ethical use of AI in warfare and surveillance.
Concerns extend beyond military applications
Mayrhofer's objections also reflected broader concerns about government power and future uses of AI.
In the resignation letter, he expressed fears that language permitting AI use for "any lawful purpose" could eventually enable forms of mass surveillance.
As a European academic, he said he was increasingly concerned about how AI technologies could be deployed in geopolitical and governmental contexts.
"This deal implies that Google (AI) products will likely be used directly against me and mine," he wrote.
The comments underscore a wider debate unfolding across the technology sector as governments seek greater access to advanced AI systems for defence, intelligence and security purposes.
For some employees, the question is no longer whether AI should be developed, but how it should be used and who should ultimately control it.
A larger debate for the AI industry
Mayrhofer will remain at Google through his contractual notice period, which runs until the end of August, according to the resignation letter. He also said he would immediately distance himself from any work connected to AI systems that could fall under the defence agreement.
His departure arrives at a time when AI companies face increasing pressure to balance commercial opportunities, government partnerships and employee expectations.
As artificial intelligence becomes more deeply embedded in military, intelligence and public-sector operations, technology firms are likely to encounter more difficult questions about ethics, governance and employee consent.
For Google, the resignation represents more than the departure of a senior security leader.
It highlights a growing fault line within the AI industry itself: whether technological capability should determine what companies build, or whether ethical boundaries should continue to shape where that technology is ultimately deployed.