In 2017 the United States Department of Defense established the Algorithmic Warfare Cross-Functional Team, known as Project Maven, to apply computer vision to the flood of imagery collected by military drones. The April 25, 2017 memo that created the team, signed by the Deputy Secretary of Defense and published on the DoD CIO’s site, set the team the task of turning that imagery into usable intelligence faster than human analysts could. To do the machine-learning work, the Pentagon contracted with technology companies - and one of them was Google. That much is on the first-party record: Maven existed, it was about applying AI to drone surveillance imagery, and it needed outside technical partners.
The internal reaction at Google is the part of the story that lives mostly in reporting rather than in first-party documents. According to contemporaneous coverage, thousands of Google employees signed a letter to CEO Sundar Pichai arguing that “Google should not be in the business of war” and demanding the company drop the contract, and a number of employees reportedly resigned in protest. The letter circulated internally and was described by journalists who obtained it; it was not published by Google as a first-party document, so its exact text and signatory count are treated here as reported, not as primary fact. What can be stated from the record is the outcome: Google decided not to renew the Maven contract when it expired.
Google’s own first-party response came on June 7, 2018, when Pichai published a post titled “AI at Google: our principles.” It set out seven principles for the company’s AI work and, more pointedly, a list of applications Google said it would not pursue - among them “weapons or other technologies whose principal purpose or implementation is to cause or directly facilitate injury to people.” The post added that “while we are not developing AI for use in weapons, we will continue our work with governments and the military in many other areas.” That document is the company’s authoritative statement of where it landed, and it is the primary source for this entry’s central claim: a major AI lab, under pressure from its own workforce, drew a public line about military uses of its technology.
The episode matters because it was an early, concrete instance of a question that has only grown louder: who gets to decide how powerful AI is used, and what happens when the people building it disagree with the people selling it. The employee revolt, the lapsed contract, and the published principles together marked the moment that internal dissent over AI’s applications became a force that shaped corporate policy at one of the field’s leading companies.