Nvidia's CEO, Jensen Huang, emphasized that China is a formidable nation, and success in this domain cannot be swiftly achieved. He highlighted China's robust willpower and extraordinary capabilities, noting that half of the world's AI researchers hail from China. Additionally, Chinese researchers hold a prominent position in American AI laboratories. Consequently, the competition in the realm of artificial intelligence promises to be a drawn-out contest, with victory far from immediate.

The United States' embargo on chip exports to China has significantly alarmed NVIDIA CEO Jensen Huang, who acknowledges the formidable potential of China's domestic computing power. Recently, Huawei introduced the AI computing power cluster solution, CloudMatrix 384, which has achieved a groundbreaking performance leap through architectural innovation. Boasting a single-node computing power of 300 PFlops, this solution has been deployed on a large scale at the Wuhu data center, marking it as the first formally commercialized super-node cluster in China. This milestone signifies a significant advancement in China's AI computing capabilities.

As April drew to a close, the instant retail landscape heated up, with Alibaba elevating its Taobao "Hourly Delivery" service to "Taobao Flash Buy" within the Taobao and Tmall ecosystem. Initially rolled out in over 50 cities, the service is set to expand nationwide on May 6. This strategic enhancement aims to consolidate internal resources, adapt to industry shifts, and bolster Alibaba's market presence.

Li Qiuseo, Chief Domestic Strategy Analyst of the Research Department at China International Capital Corporation (CICC), highlighted that the revaluation of Chinese assets is an ongoing trend. Despite the substantial global asset volatility sparked by the recent US 'reciprocal tariffs' policy, which even eclipsed the trade friction period from 2018 to 2019, Chinese assets have demonstrated greater stability and outperformed their US counterparts. Li stressed that the current landscape presents more opportunities than risks for Chinese assets and anticipated a gradual uptick in overseas investors' interest in these assets. Furthermore, he expressed confidence that A-shares might perform even better in the second half of the year compared to the first half.

Tech stocks across the board experienced a significant surge, with notable gains including Xiaomi Group rising over 6%, Alibaba up 3.8%, Tencent climbing 2.56%, and Kingsoft Cloud soaring more than 4%.

During the first quarter of this year, Guangxi has successfully inked contracts for 24 'AI +' industrial projects, amounting to a total investment of 19 billion yuan. Among these projects are the Liuzhou ASEAN Smart Computing Center, Laibin Big Data Center and Energy Storage Power Station, Guilin Cloud Smart Manufacturing Robotics Headquarters and R&D Base, Liuzhou Smart Robotics Production Base, Nanning Autonomous Vehicle Smart Manufacturing, and numerous others. These groundbreaking AI product and service projects are currently being rolled out across Guangxi, heralding a new era of technological advancement in the region.

Anthropic has unveiled substantial enhancements to its AI assistant Claude, introducing two groundbreaking features: the "Integrations" framework and the "Advanced Research" mode. The "Integrations" framework seamlessly connects with mainstream tools, streamlining task management and data integration processes. Additionally, the "Advanced Research" mode dramatically accelerates the speed of information search and processing. Currently, these two features are available for testing to a select group of users and will be progressively rolled out, with the goal of empowering users to complete tasks more efficiently.

Several open-source projects on GitHub boast the ability to acquire and decrypt local data from the WeChat client. This decrypted data is then utilized to develop AI avatars, perform data analysis, and for other applications. Nonetheless, the WeChat Security Center has recently issued a stern warning, announcing measures to clamp down on such practices in order to safeguard the security and privacy of user data. Users are strongly advised to refrain from employing third-party tools that may pose a risk of data leakage.

In a recent earnings call, Apple CEO Tim Cook disclosed that the personalized Siri project has achieved notable milestones and is anticipated to debut within the next year. This enhanced Siri will boast capabilities such as personalized context understanding, in-app operations, and screen perception. Furthermore, iOS 18 will usher in intelligent features like advanced writing tools and Genmoji, alongside seamless integration with ChatGPT, AI-powered photo search, and smart replies. Cook emphasized Apple's unwavering commitment to high standards, underscoring that new features are released only after they have reached a state of maturity.

The debate over the United States' "strictest AI chip ban" is heating up. Recently, Anthropic, a U.S.-based AI startup and developer of Claude—a competitor to ChatGPT—submitted comments to the U.S. Department of Commerce regarding the interim final rule of the Artificial Intelligence Diffusion Framework. In these comments, Anthropic urged for stricter export controls on high-end AI chips. However, what has stirred up even more controversy is Anthropic's official blog post, which accused China of establishing elaborate smuggling networks for chips, involving sums of money potentially reaching hundreds of millions of U.S. dollars.

Google has launched a dedicated AI Mode feature tab, positioned as a stand-alone search option alongside news, images, and other categories. Leveraging the advanced Gemini 2.0 model, this feature offers AI-generated direct answers, capable of multimodal processing to delve deeper into related information. Unlike the traditional search results list, AI Mode prioritizes the display of comprehensive answers. It also introduces innovative features such as location and product cards, along with search history, all aimed at enriching the user's search experience. Currently, this feature is undergoing testing with a select group of users in the US, with plans to gradually expand its reach in the future.

DeepSeek has unveiled its latest model, DeepSeek-Prover-V2, which is dedicated to mathematical theorem proving and has achieved groundbreaking results across several high-difficulty benchmarks. In the prestigious Putnam test, DeepSeek-Prover-V2 solved an impressive 49 problems, vastly outperforming the current leader, Kimina-Prover, which managed only 10 problems. Kimina-Prover is a collaborative effort between Kimi and the esteemed AIME2024 champion team Numina. The exceptional performance of DeepSeek-Prover-V2 underscores its formidable prowess in the realm of mathematical theorem proving, even with a relatively compact 7B parameter count, showcasing skills that elude much larger 671B models.

China's National Internet Security Emergency Response Center has recently uncovered a significant number of malicious foreign websites and IP addresses, predominantly originating from countries like the United States, Sweden, and India. These entities are being weaponized by foreign hacker groups to launch cyberattacks against China and various other nations. The tactics employed include setting up botnets, exploiting backdoors, and stealing confidential information, thereby posing a grave threat to Chinese network entities and individual users.

Apple's CEO, Tim Cook, has shared an update on the progress of Siri's highly anticipated personalized features. While acknowledging a slight delay, Cook assured that significant strides have been made and that these features are now on track for a release within the next year. Moreover, Apple has introduced a suite of new smart tools within the iOS system, encompassing a writing assistant, Genmoji emoticons, and seamless integration with ChatGPT. This integration aims to elevate users' daily experiences across various digital interactions. Cook emphasized Apple's commitment to maintaining rigorous standards, ensuring that all new features are thoroughly tested for maturity and stability before they reach the market. The forthcoming personalized Siri will boast groundbreaking capabilities, including contextual understanding and in-app operations, promising an even more refined and intuitive user experience.

During the fiscal 2025 Q2 earnings call, Apple CEO Tim Cook acknowledged that the rollout of Siri powered by Apple Intelligence has encountered a delay. Cook emphasized that Apple requires additional time to ensure Siri meets rigorous quality standards, but he assured that the project is currently advancing seamlessly.

Microsoft is currently engaged in negotiations with xAI to host the Grok AI model, aiming to make it available to both customers and internal product teams via its Azure cloud service. Should the agreement be finalized, Grok will be integrated into Microsoft's Azure AI Foundry platform, which grants developers access to AI services, tools, and pre-built models, facilitating the development of AI applications and agents. This strategic move will empower developers to seamlessly integrate Grok into their applications, while also opening up the possibility for Microsoft to leverage the AI model within its own suite of applications and services.

Microsoft shares have surged by over 10%, reaching a three-month high and surpassing Apple with a market value of $3.23 trillion to reclaim the title of the world's most valuable company. This impressive achievement is attributed to the company's robust financial report and promising growth expectations for its Azure cloud business.

Google has announced that it will publicly unveil its AI-driven search engine tool for the first time, with plans to introduce the "AI Mode" feature to a select group of American users within the coming weeks. This development will allow users to trial the search-focused chatbot beyond Google's lab setting, enhancing their search experience with greater intelligence and convenience.

On May 1st at 10:23 AM, OpenAI's founder, Altman, announced the retirement of GPT-4, hailing it as a groundbreaking product and noting that its data would be archived on dedicated hard drives for future historians to examine.

Li Yanhong has voiced criticism towards DeepSeek, pointing out its notable issues with hallucination. Indeed, the phenomenon of hallucination in large-scale models presents a multifaceted challenge. Emerging as a domestic open-source large model earlier this year, DeepSeek-R1 briefly rose to prominence, surpassing ChatGPT on Apple's US App Store free download rankings. This feat was attributed to its robust reasoning capabilities and writing style that resonates with Chinese expression habits, momentarily becoming synonymous with AI excellence.