Firms like OpenAI and Meta claim a compute shortage while also exploring selling compute capacity. This isn't a contradiction but a strategic evolution. They are buying all available supply to secure their own needs and then arbitraging the excess, effectively becoming smaller-scale cloud providers for AI.

Cerebras's IPO pricing reveals extreme valuations in AI hardware. At a potential 70 times its current revenue run-rate (not profit), investors are betting on hyper-growth where today's sales are a rounding error compared to future demand for specialized AI chips. This reflects a belief that compute demand will continue to grow exponentially.

AI companies with the foresight to sign long-term, multi-year compute contracts gain a significant margin advantage. They lock in prices based on past valuations, while competitors are forced to buy capacity at much higher current market rates driven up by the increasing value of new AI models.

Greg Brockman simplifies OpenAI's business to its most fundamental level: buying or building massive amounts of compute and reselling it with an intelligence layer on top. This framing reveals that their primary growth vector and constraint is access to computation, making their core operation a margin-based resale of processing power.

To secure a foundational customer like OpenAI, capital-intensive infrastructure startups like Cerebrus may have to offer extremely generous terms, including massive, near-free equity stakes. This "deal they had to take" dynamic is necessary to overcome the cold start problem and achieve scale, demonstrating the immense leverage held by large AI model companies.

OpenAI leveraged its massive demand for compute to secure warrants for a potential 11% stake in chipmaker Cerebrus for a fraction of a penny per share. This deal, tied to a $20 billion multi-year purchase commitment, highlights the immense bargaining power held by major AI model developers over their supply chain.

Despite its high valuation post-IPO, AI chipmaker Cerebras's long-term strategy focuses on inference, not just training. The bet is that inference will become a much larger segment of the AI compute market. By developing chips specifically optimized for this task, Cerebras aims to take significant market share from NVIDIA.

During major technology shifts like the move to cloud or AI, the best companies (e.g., hyperscalers, Snowflake) often have terrible early margins. In AI, inference costs are falling so rapidly that a company's margin profile can improve dramatically. Judging an early AI company on SaaS-era margin expectations is a mistake.

While training has been the focus, user experience and revenue happen at inference. OpenAI's massive deal with chip startup Cerebrus is for faster inference, showing that response time is a critical competitive vector that determines if AI becomes utility infrastructure or remains a novelty.