Reporter | Ke Yang | Editor | Zhang Yiming
Thursday, July 9th, 2026
NO.1Some Chinese AI models are 90% cheaper than those in the United States
Recently, the adoption rate of Chinese models in American enterprises has been increasing due to the continuous rise in model pricing by top AI manufacturers in the United States. Because compared to expensive American AI models, AI models from China have performance close to top American cutting-edge models and are priced lower, gradually gaining favor from American companies.
Industry insiders point out that currently, the performance of some top open source and open weight models in China is about 6 to 9 months behind the top cutting-edge models in the United States such as OpenAI and Anthropic, while the pricing is 60% to 90% lower, and can cover the vast majority of routine AI tasks, making them highly favored by American companies.
Comment:From a broader perspective, the penetration of Chinese AI in the US market reveals a emerging global AI landscape: the efficiency driven development path is gaining market recognition. This is no longer a parameter competition in the laboratory, but a game of cost and utility in real scenarios.
NO.2 Ministry of Industry and Information Technology issues warning on preventing Claude Code backdoor risks
On July 8th, the Network Security Threat and Vulnerability Information Sharing Platform of the Ministry of Industry and Information Technology issued a risk warning on preventing the hidden dangers of the Claude Code security backdoor in AI programming tools. Recently, the Network Security Threat and Vulnerability Information Sharing Platform (NVDB) of the Ministry of Industry and Information Technology discovered that the AI programming tool Claude Code has hidden security backdoors, posing a serious threat. Claude Code is an AI programming tool developed by Anthropic in the United States, which can independently complete code writing, repair, and other tasks according to textual requirements. Due to its built-in monitoring mechanism, sensitive information such as user location and identity can be transmitted back to remote servers without user consent. The affected Claude codes are versions 2.1.91 to 2.1.196.
Comment:The timely release of risk warnings by the Ministry of Industry and Information Technology, requiring relevant units and users to immediately investigate, uninstall or upgrade, and strengthen traffic monitoring, is undoubtedly a necessary protection of user rights.
NO.3 Momenta, the first physical AI stock, has landed on the Hong Kong stock market
On July 8th, Momenta, the first physical AI stock, officially landed on the Hong Kong Stock Exchange. It is worth mentioning that in terms of the cornerstone investor lineup, GIC、 Top players in the industry such as Fidelity International, Mercedes Benz, and Huaxia Fund have teamed up to make an appearance. Industry analysts interpret that Momenta's growth potential in the coming years has been collectively recognized by long-term funds, industry leaders, and other "roles".
Comment:The really interesting thing about Momenta's listing is not the addition of an AI concept stock, but the market's recognition that allowing AI to understand the physical world and act safely is worth a separate valuation system.
Disclaimer: The content and data in this article are for reference only and do not constitute investment advice. Please verify before use. Based on this operation, the risk is borne by oneself.

Li Bin commented on the "fallback" service of intelligent driving: there is no need to issue a special statement, as NIO's service standards will not fail to provide a fallback

Technology Show | National Computing Power Base Completed, AI Infrastructure Enters 100000 Card Era

People "collaborates with Yanggu Health to document the" balance "journey of this group of people