Microsoft, one of the world’s leading technology giants, has taken the unprecedented step of suspending certain cloud and artificial intelligence (AI) services to a division of Israel’s Ministry of Defense. The move, announced on Thursday, September 25, 2025, follows an urgent internal and external review triggered by a series of investigative reports that exposed the Israeli army’s use of Microsoft’s Azure cloud platform to store vast quantities of surveillance data on Palestinian civilians in the West Bank and Gaza.
The revelations, first published in August 2025 by +972 Magazine, Local Call, and The Guardian, detailed how Unit 8200, Israel’s elite cyber warfare agency, was intercepting and storing millions of mobile phone calls from Palestinians on Microsoft’s servers. According to The Guardian, this trove of data—one of the world’s most intrusive collections of surveillance information targeting a single population—was used over the past two years to plan lethal airstrikes in Gaza and facilitate arrests in the West Bank.
Microsoft’s decision marks the first time a major U.S. technology company has revoked access to its products for the Israeli military since the escalation of the Gaza conflict nearly two years ago. While the company has not disclosed every detail of the services it disabled, it confirmed that both cloud storage and AI subscriptions related to the surveillance project were suspended. In a letter to Israel’s Defense Ministry, Microsoft made clear that the decision came after it “identified evidence” supporting elements of the media reporting and determined that its terms of service had been violated.
Brad Smith, Microsoft’s president and vice chair, communicated the company’s stance in an email to employees: “We do not provide technology to facilitate mass surveillance of civilians. We have applied this principle in every country around the world, and we have insisted on it repeatedly for more than two decades.” Smith emphasized that Microsoft’s commitment to privacy protection is not just a legal or ethical obligation, but a core business value: “As employees, we all have a shared interest in privacy protection, given the business value it creates by ensuring our customers can rely on our services with rock solid trust.”
The company’s review, which is ongoing, was spurred by mounting internal dissent and public outcry. Employee protests erupted at Microsoft’s Redmond, Washington headquarters after the exposé, with some staffers staging sit-ins and hanging banners in Smith’s office. Five employees who participated in these actions were fired, a move Microsoft said was due to “serious breaches of company policies” and “significant safety concerns.” Despite these terminations, the pressure from within the company was unmistakable and contributed to the urgency of Microsoft’s investigation.
According to Reuters and CNN, the surveillance project at the heart of the controversy began in November 2021, when Yossi Sariel, then-commander of Unit 8200, traveled to Microsoft’s Seattle headquarters to meet with CEO Satya Nadella. Sariel proposed storing up to 70 percent of his agency’s data—including highly classified intelligence—on Azure servers. Prior to this partnership, Unit 8200’s internal servers could only hold recordings of tens of thousands of phone calls from Palestinians deemed “suspects.” With Azure’s near-limitless capacity, the unit’s ambitions soared: internal sources described the goal as storing “a million calls an hour.”
This vast collection of intercepted communications enabled the Israeli military to gather potentially incriminating information on virtually any Palestinian in the West Bank or Gaza. As reported by +972 Magazine, such data could be used for blackmail, administrative detention, or even to retroactively justify military killings. The scale and scope of the surveillance drew sharp criticism from human rights organizations, with a growing number determining that Israel’s conduct in Gaza amounts to genocide—a finding echoed by a United Nations commission just a week before Microsoft’s announcement.
Microsoft’s initial response in May 2025 was more cautious. Following an earlier investigation, the company stated that it had found no evidence its technology was being used to harm Palestinians. However, the August exposé, along with pressure from activists and employees, prompted an “urgent” external review led by the U.S. law firm Covington & Burling. This review uncovered that the Israeli Ministry of Defense had consumed significant Azure storage capacity in the Netherlands and made extensive use of Microsoft’s AI services, directly supporting the media’s findings.
Within days of the exposé’s publication, Unit 8200 reportedly began transferring its surveillance data out of Microsoft’s servers abroad. According to intelligence sources cited by The Guardian, the unit planned to migrate the trove—amounting to some 11,500 terabytes, or about 195 million hours of audio—to Amazon Web Services (AWS). AWS, which has become increasingly integral to Israeli military operations since October 7, did not immediately respond to requests for comment. Other tech firms, including Google and Oracle, continue to provide cloud infrastructure to the Israeli government; Google, for instance, maintains a $1.2 billion contract with Israel’s military and government, and its free services have been used in facial recognition programs in Gaza, according to The New York Times.
Notably, Microsoft’s action is limited in scope: while the company has disabled specific cloud and AI services for the surveillance project, it continues to support Israel’s cybersecurity efforts and maintains longstanding business relationships with other Israeli military units. In fact, during the most intense phases of Israel’s aerial campaign in Gaza, Microsoft’s sales of AI services to the Defense Ministry rose significantly, as revealed by leaked documents reported by +972 Magazine and The Guardian.
The Israeli Ministry of Defense has not commented publicly on Microsoft’s decision. An official, speaking anonymously to The New York Times, confirmed that Microsoft had cut off access to cloud storage services used for Palestinian phone logs but noted that Israel had already started moving some services to alternative platforms, including AWS, before Microsoft’s review concluded.
Human rights advocates welcomed Microsoft’s move as a potential turning point. Matt Mahmoudi, a researcher with Amnesty International, called the decision “a wake-up call to other companies supplying cloud infrastructure, A.I. and surveillance products to Israel.” The hope among activists is that Microsoft’s action will set a precedent for greater corporate accountability in the technology sector—especially when it comes to the ethical implications of powerful digital tools in conflict zones.
For now, Microsoft’s decision stands as a rare example of a tech giant drawing a line in the sand, asserting that its platforms should not be used to enable mass surveillance of civilian populations. As the review continues and the world watches, the broader industry faces mounting pressure to examine its own role in one of the most contentious and technologically sophisticated conflicts of the twenty-first century.
