Microsoft expands human rights oversight after investigation into Israeli defense ministry's use of AI
Final review details evidence, internal inquiries, and lessons from a controversy that roiled the company.
Microsoft said an external investigation found evidence supporting elements of allegations concerning the Israeli Ministry of Defense’s use of its cloud and artificial intelligence technologies during the war in Gaza, prompting the company to disable specific services and launch new oversight measures.
The disclosure comes in a final update published by the technology giant summarizing its response to concerns raised in 2025 about the alleged use of Microsoft Azure and AI products by the Israeli Ministry of Defense (IMOD). The report also outlines broader changes Microsoft plans to implement following months of internal reviews and external scrutiny.
The issue emerged after reports alleging that Microsoft technologies were being used in connection with Israeli military operations during the conflict in Gaza. The company said it received inquiries and concerns from employees, shareholders, and members of the public, leading it to initiate a series of investigations.
Microsoft first conducted an internal attorney-directed review and later commissioned a second investigation following reporting by The Guardian in August 2025. The newspaper alleged that a unit within the Israeli Ministry of Defense used Azure services to store recordings of phone calls obtained through broad or mass surveillance of Palestinian civilians in Gaza and the West Bank.
To examine the allegations, Microsoft retained the law firm Covington & Burling LLP, supported by an outside technical consulting firm operating under the law firm's direction. Microsoft said the investigation was conducted without accessing customer content, citing both customer privacy protections and the company’s position that its technologies should not be used to facilitate mass surveillance of civilians.
According to Microsoft, the investigation found evidence supporting elements of The Guardian’s reporting. The company said the findings included information relating to the Israeli ministry’s use of Azure storage services located in the Netherlands as well as its use of AI services.
Microsoft noted that Covington completed its work after the company had publicly discussed the matter in September 2025 and that the investigation’s factual conclusions remained unchanged.
Following the review, Microsoft informed the Israeli Ministry of Defense that it would cease and disable specified subscriptions and services, including certain cloud storage and AI technologies. The company said it gave ministry representatives an opportunity to provide additional information demonstrating that the use of the services complied with Microsoft’s terms and conditions.
According to Microsoft, public reporting at the time indicated that the ministry planned to move the data in question to another cloud provider.
The company emphasized that its relationship with the Israeli Ministry of Defense is based on standard commercial agreements and that all customers are subject to Microsoft's Acceptable Use Policy and Enterprise AI Services Code of Conduct. Those policies prohibit certain uses of Microsoft’s services, including activities that facilitate the mass surveillance of civilian populations.
Microsoft also acknowledged that it provides a range of technologies to the Israeli government, including software, cloud services, AI services, professional services, and language-translation tools. Like many technology companies working with governments worldwide, Microsoft said it also cooperates with Israel on cybersecurity matters.
Beyond the specific findings, Microsoft framed the episode as part of its broader human-rights due-diligence process. The company said it follows frameworks including the United Nations Guiding Principles on Business and Human Rights and conducts regular risk assessments, stakeholder consultations, and human-rights impact reviews across its business.
As a result of the investigation, Microsoft said it will implement recommendations provided by Covington and introduce additional measures aimed at strengthening governance, oversight, and clarity around the use of its technologies.
