Regional

AI News Brief: Nvidia GPU server prices are too high, driving up infrastructure costs for Japanese AI companies and pushing the industry to look for cheaper alternatives. There are two main directions right now: first, adopting NPU (Neural Processing Unit) semiconductor chips to replace or complement GPUs — for example, the NPU developed by Korean chip startup Rebellions is already being used in some AI data center servers; second, offloading compute and data access load through cache servers instead of simply piling on more GPUs.

AI News Brief: Nvidia CEO Jensen Huang is visiting Japan, with an expected event to promote semiconductor supply and support measures aimed at Noetra — an AI developer backed by the Japanese government and co-formed by SoftBank and other partners. Huang is expected to meet with the head of this government-backed AI developer. Nvidia has been partnering with sovereign AI developers across Europe, the Middle East, and elsewhere in recent years; Japan is the next target in the same strategic playbook.

AI news brief: Japan’s eight major power companies are jointly building or expanding 30 substations nationwide to strengthen transmission capacity ahead of time, in response to a new wave of power demand growth driven by AI data centers. One specific case is Kansai Transmission and Distribution, a subsidiary of Kansai Electric Power, which will expand substations in Kobe and Osaka. This round of investment is part of the power industry’s infrastructure play to get ahead of AI data center demand, expanding substation counts and transmission networks to reserve enough power capacity for future tech companies building data centers locally…

AI News Brief: Australian Prime Minister Albanese announced on Wednesday that Australia will launch ‘world-leading’ AI standards governing data center construction and model training, continuing the country’s earlier lead in legislating to ban minors from social media. Speaking at the University of Sydney in an address titled ‘AI in Australia’s Interests’, Albanese said the goal is to ’lead in shaping AI’s future, rather than being led by AI’s future’…

AI News Brief: Short news related to Japan’s nuclear power, TSMC supply chain, and AI server manufacturing — this one falls under the latter category, with limited detail. Summary below: Nikkei Asia reports that several Taiwanese manufacturers involved in producing AI data center servers are integrating AI technology into their own production workflows to accelerate shipments and meet surging market demand for higher-performance systems. The article uses Foxconn (鴻海) as an example, noting that its server rack trays for Nvidia’s next-generation Vera Rubin platform have achieved a rate of one tray per minute via fully automated production lines…

AI News Brief: Panasonic (formerly Matsushita Electric) was once the flagship enterprise of Japan’s home appliance industry. As early as two months before China launched its reform and opening-up policy in 1978, then-Vice Premier Deng Xiaoping visited Panasonic’s color TV factory in the suburbs of Osaka — a powerful symbol of Japan’s home appliance manufacturing prowess at the time and a model for China to learn from. But the home appliance market has shifted dramatically since then. Chinese and Korean manufacturers now dominate the global home appliance market, forcing traditional Japanese consumer electronics giants like Panasonic to transform in order to survive. The report notes that Panasonic has moved away from its old focus on home living products and is now developing data center-related businesses, reflecting how Japan’s old-school electronics makers are trying to find new growth momentum by shifting from hardware manufacturing into digital infrastructure. The original summary doesn’t go deeper on the specific scale, technical details, or financial figures of Panasonic’s data center business — see the original link for more.

AI News Brief: Nvidia and Mitsubishi Heavy Industries are in talks for the Japanese industrial group to supply cooling systems and energy management equipment for Nvidia’s AI data centers. According to Nikkei, Nvidia is evaluating adopting MHI’s cooling and energy management systems in its AI factories to help improve overall energy use efficiency. The partnership is still in the negotiation stage, and the original summary does not provide specific contract size, timeline, or technical specs…

AI News Brief: SoftBank Group founder and chairman Masayoshi Son stated at the SoftBank World 2026 conference in Tokyo on July 14, 2026, that global AI infrastructure will require $5 trillion in annual investment by 2040. He noted that the expansion of data centers, power supply, and humanoid robots will support AI’s shift from a human-centered work model to a new labor structure. Son expects AI to account for 20% of global GDP by then, with AI agents and humanoid robots becoming the key forces shaping the future workforce.

AI News Brief: China’s trade performed strongly in the first half of 2026, with total trade volume up 21% year-over-year, driven by a global investment boom in AI-related industries, which successfully offset both geopolitical uncertainty and weak domestic household consumption. June exports rose 27% YoY, while imports surged 36% on rising chip prices. Container throughput at Shanghai’s Yangshan Port shows that global demand for AI-related supply chains (such as chips and server components) continues to support China’s overall export momentum.

AI News Flash: Nikkei Asia’s latest Tech Latest Podcast features Taipei tech correspondent Lauly Li in conversation with host Katey Creel. The core topic: why the AI wave is triggering a full CPU revival. For a long time, AI training and inference workloads have been almost entirely dominated by GPUs, but as model sizes continue to balloon, GPU supply bottlenecks and high power consumption are getting increasingly severe. Chipmakers are actively reassessing the role of CPU in AI infrastructure—especially on the inference side—where CPUs offer cost and flexibility advantages in latency-sensitive, small-batch scenarios. Some vendors are also experimenting with heterogeneous architectures pairing CPUs with AI accelerators to diversify their over-reliance on a single GPU.