According to a Reuters report, Chinese tech giant Alibaba is planning to ban Claude Code — an AI coding tool developed by Anthropic — from internal use. The stated reason is concern over so-called 'backdoor risks,' meaning the possibility that the tool could secretly send data to outside parties. At this point, no technical evidence has been made public, and an official response from Anthropic has not been confirmed. The move is drawing wide attention as a new point of geopolitical tension in the US-China rivalry over AI tools.

Product growth expert Elena Verna criticizes what she calls 'AI confidence theater' — a pattern where AI products assert they can do things with more certainty than is accurate. Because AI systems often present answers in a confident tone even when they are wrong, users find it hard to recognize the tool's real limits. Verna encourages product designers to build more honest UX that communicates uncertainty clearly. This raises an important question for both developers and users: how should trust in AI be designed?

The okTurtles blog introduces a method called the 'short leash' approach to AI coding. Instead of handing large tasks to an AI all at once, the idea is to break work into small steps and review the output at each stage. This helps catch errors early and maintain quality throughout the process. The author draws on a real example from building a game called Fable to show a concrete workflow that programmers can start using today. The post adds a useful perspective to ongoing debates about how to design human oversight in AI-assisted development.

On July 3rd, Japan's long-term interest rate — based on the yield of 10-year government bonds — temporarily rose to 2.81%, the highest level in approximately 29 years. According to NHK, the rise was driven by growing expectations of faster inflation and concerns about the country's fiscal situation, which led investors to sell government bonds. A rise in long-term rates can affect mortgage rates and borrowing costs for businesses, so the potential impact on households and companies is being closely watched. This article aims to explain the economic context and does not recommend any specific financial action.

Japanese semiconductor giant Kioxia has announced the start of sample shipments of its latest flash memory chips from its factory in Iwate Prefecture, aiming to meet the surging demand driven by the spread of generative AI. According to NHK, the factory began operations last year, and full-scale mass production is planned for next year. Since generative AI requires large amounts of high-performance memory in data centers, this move is a clear example of Japan's semiconductor industry beginning to benefit from the global AI boom. It illustrates the growing intersection between domestic manufacturing and AI technology.