- Microsoft, after a review, cuts off some services to a unit within Israel’s Ministry of Defence over alleged surveillance.
- The move is limited in scope but seen as significant given growing scrutiny over ethics in cloud and AI use.
What happened: Microsoft disables specific Azure cloud and AI services for an Israeli defence unit following findings of misuse for mass surveillance of Palestinians
Microsoft has disabled a set of its cloud and AI services used by a unit within Israel’s Ministry of Defence. The change follows an internal review that confirmed parts of reports from The Guardian, +972 Magazine and Local Call that Microsoft’s Azure infrastructure was being used by that unit to store and analyse large volumes of intercepted civilian communications. Brad Smith, Microsoft’s vice chair and president, said the company “ceased and disabled” certain subscriptions after finding “evidence that supports elements of The Guardian’s reporting.”
The action applies only to specific services and does not sever all ties with Israel’s defence sector; cybersecurity services, for example, are maintained. The affected unit is reported to be Israel’s Unit 8200, the signals intelligence agency. Much of the data under question was allegedly stored in European data centres, including in the Netherlands. Microsoft stressed it has not accessed customer content during the investigation, basing its findings on internal data such as business records and correspondence.
Also read: Microsoft signs $17.4B GPU deal with Nebius
Also read: Microsoft seals US government AI agreement
Why it is important
This move by Microsoft marks a rare instance when a major cloud provider publicly restricts services to a military client over ethical concerns. It reinforces that tech companies are under increasing pressure not only to comply with laws like GDPR and privacy norms but also with their own internal policies regarding acceptable use of cloud and AI. For organizations that rely heavily on cloud infrastructure, this serves as a reminder to ensure contractual clarity, oversight, and ethical compliance.
From a geopolitical standpoint, disabling only some services while maintaining others signals careful calibration: Microsoft is drawing a line without a blanket withdrawal, perhaps to manage operational, legal and reputational risk. Civil society and rights groups have praised the decision but argue it should go further. The episode contributes to a growing global debate about how cloud-AI providers should manage customer conduct, especially in conflict zones, and could set precedents for future action by other tech firms.
