US orders DeepSeek out of civilian federal IT and purchasing
Aerial view of the St. Elizabeths West Campus in southeast Washington, the Department of Homeland Security headquarters site, in 2015: red-brick former hospital buildings among lawns and trees, with the green-roofed Coast Guard headquarters at right and a river in the distance..Photo: U.S. Coast Guard / Wikimedia Commons, Public domain
The Department of Homeland Security (DHS) published an order in the Federal Register on 20 October that bars products and services of DeepSeek, an AI company in China, and its affiliates from civilian federal agency procurement. Agencies must also remove DeepSeek products, and third-party products built using DeepSeek, from their IT, except national security and secure compartmented information systems. Exclusion and Removal Order #2026-01, issued under the Federal Acquisition Supply Chain Security Act of 2018 and signed by Homeland Security Secretary Markwayne Mullin, was first released on 24 July 2026. It cites DeepSeek servers in China and the possible transfer of sensitive data there.
The General Services Administration (GSA) has signed a OneGov agreement with US AI. Nextgov reports that US AI is a North Bethesda, Maryland, company and that it is GSA's first direct-to-government OneGov deal with an 8(a) certified small business. GSA's listing offers federal agencies up to 30 per cent off eligible software licences, plus two-week proof-of-value pilots for $1, on a FedRAMP High-authorised platform. GSA describes the offer as a limited-time offering that expires on 30 September 2028. The entry shows no signing date. GSA lists these terms under the US AI entry.
Governor Abigail Spanberger signed Executive Directive 6 on 23 October. It tells the Secretary of Public Safety and Homeland Security to set up a Cyber Consortium, which must adopt a statewide cyberattack prevention and response plan and an infrastructure-hardening plan that address AI-related risks. AI companies are to be asked to share with the Cyber Consortium the confidential briefings and reports they give other government bodies, including New York's Department of Financial Services under the RAISE Act and the European AI Office. The secretary is to aim for briefings on AI risks at least quarterly, and a tabletop exercise at least yearly. Other states will be invited to join.
Japan's Digital Agency said on 22 October that data may have leaked after a third party accessed the cloud environment of Studio, whose service ran part of its website. The agency lists five categories of potentially affected contact-form submissions, each a maximum, together at most 13,323, including up to 3,661 enquiries about its vaccination certificate app, plus 35 staff or contractor accounts. The agency says the scope is being examined and no misuse has been confirmed. Aomori Prefecture said on 21 October that data on 206 event applicants, collected by its contractor on Studio, may have leaked.
Oman has launched the second phase of its National Digital Economy Programme, for 2026–2030, a release on the Foreign Ministry's website says. It focuses on digitalising strategic sectors, expanding e-commerce and electronic payments, and accelerating artificial intelligence adoption. The release gives its targets as a contribution of 1.5 billion Omani rials to gross domestic product (GDP), 300 million rials in technology investment and 3,000 new direct jobs a year. The programme's Strategic Intelligence Platform, built with the World Economic Forum, was also launched. An official survey cited in the release puts the digital economy at 9 per cent of GDP.
