Sam Altman made news again, with reporting from the Financial Times that the OpenAI CEO is engaged in discussions with key Middle Eastern investors and the Taiwanese chip giant TSMC to launch a new chip venture to design and build semiconductors for accelerating AI workloads.
At the heart of this venture is the ambitious plan to develop and fabricate chips integral for training and building AI models, reflecting the growing importance of custom hardware in the rapidly expanding field of AI.
Sam Altman’s Chip Discussions
Sam Altman is discussing establishing a new venture to develop specialized chips for AI applications with prominent Middle Eastern investors and Taiwan Semiconductor Manufacturing Co, TSMC.
The venture’s extent and focus on manufacturing, a detail not previously disclosed, have recently come to light. The discussions remain in the early stages, with the complete list of partners and funders still unconfirmed. Altman’s initiative is driven by concerns over a potential shortage of AI-specific chips, as demand is expected to outpace production forecasts.
This approach, involving the construction and maintenance of semiconductor fabs, marks a departure from the strategy of OpenAI’s industry peers like Amazon, Google, and Microsoft, who typically design their silicon and outsource production. Building such facilities requires substantial investment, with a single plant costing tens of billions of dollars.
The discussions with G42 alone reportedly considered raising $8 to $10 billion. Altman’s urgency stems from a need to secure enough chip supply for the late 2020s, anticipating a surge in AI applications and the associated demand for computing power.
This initiative, aimed at reducing OpenAI’s dependence on Nvidia, involves collaboration with influential figures like Sheikh Tahnoon bin Zayed al-Nahyan of the UAE. The venture aims to create chips essential for training and building AI models and establish the necessary fabrication plants.
Sheikh Tahnoon, a key figure in this negotiation, oversees substantial investment funds and entities in Abu Dhabi and is involved with G42, an AI company collaborating with Microsoft and OpenAI.
The news has caused much handwringing from several industry analysts who are worried that Altman’s statements indicate that he intends to move OpenAI into the semiconductor business. There’s no indication, however, that OpenAI is directly involved in the chipmaking discussions. I don’t believe any potential venture would materialize as part of the AI company.
Sam Altman is an unusually active investor, even for a Silicon Valley CEO. The OpenAI CEO has funded over 400 companies through his involvement with Hydrazine Capital, Apollo Projects, and Altman Ventures. Recent reporting indicates that Altman received $75 million from the University of Michigan endowment fund to help start a new venture fund. OpenAI also operates its own AI-focused venture capital fund.
In any case, OpenAI is set to be a significant beneficiary and primary customer of the venture, aligning its hardware needs closely with its ambitious AI development roadmap.
Whether part of OpenAI or not, Altman’s moves signal a strategic shift in the AI accelerator space. Altman’s engagement with Middle Eastern investors, including prominent figures like Sheikh Tahnoon bin Zayed al-Nahyan and TSMC, the Taiwanese chipmaking giant, underscores a growing trend in the AI industry: the push towards self-reliance in semiconductor production.
This initiative, aiming to reduce dependence on NVIDIA, reflects a broader industry recognition of the need for specialized, tailored chip solutions to drive the next generation of AI advancements. This mirrors the actions the major public cloud providers took, with Amazon Web Services, Microsoft, and Google designing custom silicon solutions. Notably, the cloud providers, while designing accelerators, outsource the fabrication of the semiconductors.
The news also raises questions about how OpenAI views the offerings of NVIDIA competitors AMD and Intel. AMD recently announced its MI-300 series of AI accelerators, while Intel’s Gaudi3 is expected to be announced sometime this quarter.
A strategic move by Altman towards chip development is not just a significant development for OpenAI. Still, it is indicative of a larger trend in the AI industry toward vertical integration and hardware customization. If the venture comes to fruition, it could herald a new era of innovation and competition in both AI and semiconductor industries, with far-reaching implications for the development and application of AI technologies.