Sam Altman Reaffirms OpenAI's Commitment to Nvidia Partnership
OpenAI CEO Sam Altman has come forward to strongly defend his company's ongoing partnership with chip giant Nvidia, stating unequivocally that the artificial intelligence firm "loves working" with the semiconductor manufacturer. In a public post on social media platform X, Altman described Nvidia's processing units as "the best AI chips in the world" and expressed OpenAI's desire to remain a "gigantic customer" of the chip maker for the foreseeable future.
Market Speculation and Investment Uncertainty
Altman's comments arrive amidst growing market speculation regarding OpenAI's long-term computing strategy and reports suggesting that Nvidia might be reassessing the pace of its investments connected to the AI company. The OpenAI chief appeared to address circulating rumors directly, stating "I don't get where all this insanity is coming from" in reference to the speculation surrounding the partnership.
The backdrop to this situation involves a significant proposed investment that was announced in September 2025. At that time, Nvidia revealed plans to make a monumental $100 billion investment in ChatGPT-maker OpenAI, with both companies describing the arrangement as "the largest computing project in history" through a joint announcement featuring OpenAI President Greg Brockman and Nvidia CEO Jensen Huang.
Stalled Negotiations and Internal Doubts
Although the deal was initially expected to conclude within weeks, negotiations have extended for several months. Recent reports from the Wall Street Journal indicate that talks have stalled after some executives within Nvidia expressed reservations about the arrangement. According to sources familiar with the matter, Huang has privately pointed to increasing competition from companies such as Google and Anthropic as a concern.
More specifically, Huang has reportedly criticized what he perceives as a lack of discipline in OpenAI's business approach and raised alarms about growing competition from Google's Gemini and Anthropic's Claude AI systems. These internal doubts appear to have contributed to the prolonged negotiation process.
Contradictory Statements from Nvidia Leadership
The situation became more complex when Jensen Huang made contradictory statements about the investment during public appearances. On January 31, while speaking to reporters outside a restaurant in Taipei, Huang confirmed that "Yes, Nvidia will invest in OpenAI" and that "it'll be big," but he explicitly denied the $100 billion figure, stating "No, no, nothing like that."
Remarkably, just one day later on February 1, Huang executed a complete reversal, telling reporters that "It was never a commitment" and explaining that OpenAI had invited Nvidia to invest up to $100 billion. He characterized the invitation as an honor but emphasized that any investment would proceed "one step at a time." Despite this backtracking, Huang dismissed reports of tension between the two companies as "nonsense" and reaffirmed Nvidia's intention to make a substantial investment in OpenAI.
Technical Shifts and Competitive Dynamics
According to industry analysis, the potential friction between OpenAI and Nvidia may stem from evolving technical requirements in artificial intelligence development. While Nvidia's graphics processing units (GPUs) have proven exceptionally well-suited for the massive data processing necessary to train large AI models like ChatGPT, the focus in AI advancement is increasingly shifting toward inference and reasoning applications.
This transition to inference represents what many consider to be the next significant stage in AI development, and it has inspired OpenAI to explore alternative chip architectures. Specifically, the ChatGPT-maker has been searching for GPU alternatives since last year, with particular interest in companies developing chips that incorporate large amounts of Static Random-Access Memory (SRAM) embedded directly into the same silicon as the rest of the processor.
The technical advantage of this approach lies in the speed improvements it can offer for chatbot systems and other AI applications as they process simultaneous requests from millions of users. By integrating as much costly SRAM as possible onto each chip, companies can potentially achieve performance benefits for inference tasks.
OpenAI's Exploration of Alternative Suppliers
This strategic shift has led OpenAI to establish deals with other chip manufacturers, including Advanced Micro Devices (AMD), for GPUs designed to compete with Nvidia's offerings. Industry sources indicate that OpenAI's arrangement with AMD did not sit well with their existing partner Nvidia, potentially contributing to the current tensions between the companies.
The emergence of AI inference as a new competitive frontier, combined with OpenAI's active pursuit of alternatives in the inference chip market, represents a significant challenge to Nvidia's current dominance in the AI hardware space. This exploration of competing technologies occurs precisely as the two companies are engaged in high-stakes investment discussions, creating a complex dynamic between partnership and competition.
As the artificial intelligence industry continues its rapid expansion, the relationship between leading AI developers like OpenAI and key hardware suppliers like Nvidia will remain crucial to technological advancement. The current situation highlights the delicate balance between collaboration and competition that characterizes this cutting-edge sector of the technology industry.
