Group 1 - The core agenda of OpenAI CEO Sam Altman's visit includes ensuring stable AI server supply from Hon Hai for the "Gateway to the Stars" project and finalizing collaboration with TSMC for the production of self-developed AI chips, aiming for mass production by 2026 to reduce reliance on Nvidia [1][2][6]. - The "Gateway to the Stars" project involves building five new data centers in the U.S., marking one of the largest investments in computing power in the country, with Hon Hai being the largest AI server supplier for this initiative [4][5]. - OpenAI is actively developing its custom AI chips (ASIC) to decrease dependence on external suppliers like Nvidia, with plans to have TSMC manufacture these chips, focusing on production details for a target mass production in 2026 [6][8]. Group 2 - Altman's discussions with Hon Hai primarily focused on ensuring the production and delivery capabilities of AI servers to support the extensive infrastructure needs of the "Gateway to the Stars" project [5]. - The initial versions of OpenAI's self-developed chips are expected to be primarily used for inference rather than training, and their successful mass production will serve as a significant validation of the tech giants' self-developed chip strategies [8].

Core Viewpoint - Alibaba and Baidu are transitioning to using self-designed chips for AI model training, reducing reliance on Nvidia's chips due to increasing export restrictions from the US and government support for domestic technology [1][3]. Group 1: Company Developments - Alibaba has been using its self-developed chips for smaller AI models since the beginning of this year [3]. - Baidu is experimenting with its Kunlun P800 chip to train the new version of its Wenxin Yiyan AI model [3]. - Both companies have not completely abandoned Nvidia chips, as they still utilize them for developing their most advanced AI models [4]. Group 2: Chip Performance and Comparison - Alibaba's AI chip, Zhenwu, is reported to slightly outperform Nvidia's A100 chip released five years ago, indicating a performance gap still exists [3]. - Baidu's Kunlun P800 chip, while not as powerful as Nvidia's latest Blackwell chip, is designed specifically for large language models and can handle both inference and training tasks [3]. - Alibaba's AI chip is said to compete with Nvidia's H20, which is a scaled-down version designed for the Chinese market [3]. Group 3: Market Implications - The shift towards self-developed chips represents a significant change in China's tech and AI sectors, which previously relied heavily on Nvidia's high-performance processors [3]. - The competition in the AI chip market is intensifying, as indicated by Nvidia's acknowledgment of the emerging competition [4].