Alphabet's Google and Blackstone are planning to launch a new artificial intelligence cloud company that would use Google's proprietary AI chips to compete in the fast-growing AI infrastructure market, according to reports from Bloomberg, Reuters, and The Wall Street Journal.

Nvidia's fanless Vera Rubin AI server racks, slated for mass production in the second half of 2026, are expected to accelerate global adoption of liquid cooling across server components, creating demand beyond GPUs for CPUs, memory, network cards, and switches, and reshaping thermal management supply chains worldwide and vendor economics.

Chinese officials have renewed calls to integrate artificial intelligence (AI) more deeply into manufacturing as Beijing seeks to modernize traditional industries, strengthen industrial competitiveness, and cultivate new growth drivers.

Baidu said its core AI businesses became the majority of the company's revenue in the first quarter of 2026, driven by rapid growth in AI cloud infrastructure, expanded foundation-model support on its model-as-a-service platform Qianfan, and stronger commercial adoption of its Kunlunxin AI chips.

Analysts offered mixed assessments of Baidu after the Chinese internet company posted first-quarter revenue that topped market expectations, with some pointing to stronger AI monetization. In contrast, others warned that profitability remains under pressure from rising investment and lower-margin AI businesses.

Australia directed six China-linked investors to sell their combined 17% stake in rare earths miner Northern Minerals within two weeks, as reported by The Wall Street Journal and the Australian Broadcasting Corp. The treasury issued the order to protect national interest and to support efforts to build a rare earths supply chain outside China.

Anthropic launched Claude for Legal, a specialized version of its Claude generative AI designed to integrate directly with law firms' existing tools and workflows, the company announced. The offering arrives as legal use cases have surged on Claude, and it targets contract analysis, litigation support, employment, intellectual property, and privacy compliance workflows.

Anthropic urged US policymakers to tighten chip export controls and pass legislation banning so-called distillation attacks, warning that failure to act could allow China to shape global AI rules by 2028. In a recent policy report, the AI developer said that compute capacity remains the decisive factor in the US-China AI competition and that existing export controls have given the US a current lead.

CyCraft announced an expansion of its Japan strategy at its CyCraft Day Japan partner summit, teaming with local cybersecurity providers, including NTT Security and Future Secure Wave, to offer AI-managed security services amid rising geopolitical risks and supply chain attacks. The vendor said more than 20 Japanese cybersecurity service providers and 70 representatives from industry, government, and academia attended the event, where CyCraft outlined a shift from channel sales to local value creation in Japan.

Anthropic agreed to brief members of the Financial Stability Board, the G20 body for finance officials and central bank governors, on the cyber risk profile of its Claude Mythos model after a request from the Bank of England governor, sources told Financial Times. The sessions will explain how Mythos can automatically uncover cybersecurity vulnerabilities across global banks and financial systems and outline the model's potential impact on cyber defenses and financial stability.