OpenAI, the renowned creator of ChatGPT and DALL-E AI models, is currently considering a groundbreaking move into developing its own artificial intelligence (AI) chips. Reports also suggest that OpenAI has examined a potential acquisition opportunity as part of this strategic initiative.
🔍 According to sources familiar with the discussions within the company, OpenAI has not yet reached a final decision on this pivotal step. Nonetheless, the persistent shortage of high-cost AI chips, which OpenAI heavily relies on, has been a topic of concern for the company for some time, with discussions dating back to at least last year.
🧩 OpenAI has explored several strategies to address this challenge, including the possibility of in-house chip manufacturing, forging closer collaborations with established chipmakers like Nvidia, and diversifying its chip suppliers beyond its current dependence on Nvidia (NVDA.O).
👨💼 CEO Sam Altman has unequivocally expressed the critical importance of securing additional AI chips for OpenAI. He has publicly voiced his concerns over the scarcity of graphics processing units (GPUs), a market segment overwhelmingly dominated by Nvidia, which holds over 80% of the global GPU market share, ideally suited for AI applications.
💻 The endeavor to acquire more chips aligns with two key concerns raised by Altman: the shortage of advanced processors crucial for powering OpenAI's software and the substantial costs associated with maintaining the necessary hardware infrastructure for supporting its AI endeavors and products.
💡 Since 2020, OpenAI has been developing its generative AI technologies on an expansive supercomputer provided by Microsoft, one of its major backers, which incorporates 10,000 of Nvidia's GPUs.
💰 Operating ChatGPT is a resource-intensive process for the company. Each query costs approximately 4 cents, as estimated by Bernstein analyst Stacy Rasgon. Should ChatGPT queries scale up significantly, akin to Google's search volume, it would necessitate an initial investment of approximately $48.1 billion in GPUs and an annual expenditure of about $16 billion on chips to sustain operations.
🌟 Entering the Age of Customized Chips
Should OpenAI decide to venture into the development of its AI chips, it would join an exclusive group of prominent tech giants, including Alphabet's Google (GOOGL.O) and Amazon.com (AMZN.O), that have sought to take control of the chip design process, a fundamental aspect of their core operations.
🤔 Nevertheless, the path ahead remains uncertain as OpenAI contemplates the creation of its custom chips. Such a move would constitute a significant strategic undertaking, involving substantial investments that could reach hundreds of millions of dollars annually, as observed by industry experts. Success is far from guaranteed even with dedicated resources.
🏭 Acquiring an established chip company could potentially expedite the process of developing OpenAI's custom chip, a strategy reminiscent of Amazon.com's acquisition of Annapurna Labs in 2015. OpenAI had reached the point of conducting due diligence on a potential acquisition target, though the identity of this company remains undisclosed.
🔄 Even if OpenAI proceeds with its custom chip plans, including an acquisition, it is expected to be a multi-year endeavor, leaving the company reliant on commercial chip providers like Nvidia and Advanced Micro Devices (AMD.O) in the interim.
🌐 Notably, some major tech companies have embarked on custom chip development projects over the years, yielding mixed results. Meta's (META.O) custom chip endeavors have encountered challenges, leading to the shelving of certain AI chip projects, according to a Reuters report. Meta is now concentrating on a new chip designed to accommodate all AI workloads.
🔌 Microsoft (MSFT.O), a primary supporter of OpenAI, is also reportedly in the process of developing its custom AI chip, which OpenAI is currently testing. These developments may signify a growing divergence between the two companies.
🚀 The demand for specialized AI chips has surged since the launch of ChatGPT last year. Dedicated AI accelerators are indispensable for training and operating the latest generative AI technologies, with Nvidia emerging as one of the few chip manufacturers capable of producing such AI chips, maintaining a firm grip on the market.
🔬 OpenAI's exploration into AI chip development underscores the company's commitment to addressing the challenges posed by the AI chip shortage and advancing its position in the rapidly evolving AI landscape. The outcome of these deliberations will undoubtedly have profound implications for OpenAI's future endeavors and the broader AI industry. OpenAI has chosen not to comment on these developments.
That's sound great news