Public Policy & Governance

The Department of Defense filed a formal response to Anthropic's lawsuit defending its decision to exclude the company from warfighting applications, stating Anthropic 'cannot be trusted' with military systems due to its restrictive acceptable use policies. According to Wired, the government argued it 'lawfully penalized the company for trying to limit how its Claude AI models could be used by the military.' Separately, a defense official told MIT Technology Review that the Pentagon is planning to establish secure environments where generative AI companies can train military-specific model versions on classified data.

Meanwhile, OpenAI reportedly signed a partnership with AWS to sell AI systems to the U.S. government for both classified and unclassified work, as reported by TechCrunch, expanding beyond its existing Pentagon contract. The Trump administration stated in a Bloomberg report that it would pursue a 'legal fight to oust Anthropic PBC from all US government agencies' following the dispute. TechCrunch noted the Pentagon is actively developing alternatives to Anthropic's technology.

UK Technology Secretary Liz Kendall announced £1 billion in funding for large-scale quantum computer design, explicitly framing it as a lesson learned from the country's failure to retain AI talent and startups. According to The Guardian, Kendall stated the government 'will not let quantum computing talent slip through its fingers' and emphasized the need to prevent homegrown quantum startups, engineers, and researchers from relocating to competing nations. The funding is intended for scientists, researchers, public sector entities, and businesses working on quantum technology.

The announcement reflects growing recognition among European policymakers that early-stage technology leadership does not translate to economic capture without deliberate industrial policy. The UK has historically produced world-class AI research through institutions like DeepMind, only to see commercial value and talent migrate to the United States. Kendall's explicit reference to 'lessons from US dominance of the AI race' signals a shift toward more interventionist approaches in emerging technologies before similar patterns repeat.

Chinese AI agent platform OpenClaw has captured significant market momentum with backing from local governments encouraging AI agent creation for productivity gains, according to Financial Times. Nvidia CEO Jensen Huang called OpenClaw 'the next ChatGPT' at the company's GTC conference, as reported by Bloomberg, triggering a rise in related Chinese stocks. The platform has sparked what the FT described as 'AI lobster' fever, with the phenomenally popular agent serving as a cultural marker of China's distinct AI development trajectory.

Separately, Alibaba raised prices for AI computing and storage products by up to 34% in response to surging demand and rising infrastructure costs, according to Bloomberg. Bloomberg also reported that Tencent is seizing initiative in China's agentic AI competition, challenging Alibaba's earlier lead in rollout speed and user growth. Nvidia announced it has restarted H200 chip manufacturing for Chinese customers after receiving U.S. export licenses for 'many customers in China,' per Bloomberg.

Partnership on AI published guidance on shaping AI transparency processes with the National Institute of Standards and Technology, contributing to the technical standards development that underpins federal AI regulation. While details of the specific recommendations were not provided in the brief announcement, the timing aligns with NIST's ongoing work to operationalize the AI Risk Management Framework and establish measurement standards for AI system transparency. NIST standards typically become de facto requirements through their incorporation into procurement rules and regulatory guidance across federal agencies.

This follows a broader pattern of multi-stakeholder organizations attempting to influence the technical specifications that will govern AI deployment. The UK's Advertising Standards Authority also took regulatory action this week, as reported by BBC, banning an advertisement for an AI editing app that claimed it could 'remove anything,' stating the ad condoned 'digitally altering and exposing women's bodies without their consent.' This marks one of the first advertising enforcement actions specifically addressing AI capability claims in the context of non-consensual intimate images.