Al Jazeera reported that the government is expected to contest the ruling on the Pentagon's blacklisting of Anthropic, which followed the company's public criticism of the defence department's position on deploying AI on the battlefield. A separate designation under a different statute remains before the federal appellate court in Washington DC.
Bloomberg reported on 26 August that Huawei has submitted a formal bid to build advanced AI data centres in Egypt serving government functions including defence, public surveillance and a national civil registry database. The State Department has mobilised Nvidia, AMD and Microsoft behind a competing offer while invoking export licensing controls and warnings over prohibited Huawei hardware.
MLex reported on 25 August that Thailand's National Digital Economy and Society Commission is weighing an AI governance sandbox to test guidelines for the use of AI inside government agencies. The Digital Economy and Society Ministry would work with its supervised agencies to align public sector AI deployment with governance principles.
Regulation
The Ministry of Law of Singapore and the Intellectual Property Office of Singapore opened a public consultation on 26 August 2026, running to 22 October, on the impact of AI on the country's intellectual property regime. On copyright it seeks views on whether the scope of the computational data analysis exception is sufficiently clear, how lawful access and rights owner safeguards should operate, how responsibility should be allocated among developers, deployers and end users when generated output infringes, and how human contribution should be evidenced in AI-assisted works. On patents it asks how inventorship principles apply across the spectrum of human and machine interaction in the inventive process, and how large-scale publication of AI-generated technical disclosures may distort the prior art landscape.
The National Cabinet of Australia agreed in Sydney on 26 August 2026 to a nationally consistent regulatory framework setting minimum requirements for large data centres, to be legislated in early 2027. First Ministers accepted that large data centres carry material energy, water and land use consequences, and the Commonwealth will work with states and territories on consistent mandatory standards in those areas, including conditions attached to the delivery of AI training, with legislation designed to complement rather than duplicate state planning and approval processes.
The Ministry of Industry and Information Technology of China told a State Council press conference on 26 August 2026 that nearly 200 key standards have now been formulated for the AI sector. Vice Minister Xin Guobin said AI has been integrated across the full industrial chain from research and design through production, quality inspection and maintenance, and that multiple cities are being organised to carry out AI ethics review and service practices.
Cases
Via the ruling in Anthropic PBC v United States Department of Defense, in the United States District Court for the Northern District of California, it can be seen that Judge Rita F. Lin held on 27 August 2026 that the Pentagon's designation of the company as a national security supply chain risk was unlawful. The 59-page order found the measures constituted unlawful retaliation in breach of the First Amendment and violated Fifth Amendment due process, and that the designation was arbitrary and capricious, recording that the department remained free to select the AI vendor of its choice but that the broad measures imposed were illegal and baseless. The court relied on the government's continued dealings with the company after the designation, including an under secretary's statement that the parties were close to a deal and later White House discussion of deploying a newer model in sensitive settings, as inconsistent with a genuine fear of sabotage, and rejected the argument that speech within contract negotiations falls outside constitutional protection.
Academia
arXiv hosts 'Mapping General-Purpose AI Governance in Twenty AI Middle-Power Jurisdictions' by Josephine Schwab, Nathan Naidoo, Ferruccio Barazzutti, Sheryn Lee and Caio Vieira Machado, submitted on 18 August 2026. Mapping twenty jurisdictions at provision level, the authors find that roughly one provision in five sits in binding law, three quarters of binding instruments do not define general purpose AI, and outside the EU no jurisdiction imposes a binding evaluation duty on a model developer.
Events
The John L. Weinberg Center for Corporate Governance at the University of Delaware will hold AI, Agentic Companies and the Future of Corporate Governance on 6 October 2026 at Clayton Hall in Newark, Delaware. The programme examines how existing legal and governance frameworks apply to organisations relying on autonomous AI systems, including directors' duties, individuation and liability for AI agents, and regulatory approaches to AI-enabled entities.
Sources: Al Jazeera, Bloomberg, MLex, Ministry of Law of Singapore, National Cabinet of Australia, Ministry of Industry and Information Technology, Govinfo.gov, arXiv, Weinberg Center for Corporate Governance