Developing custom chips could place OpenAI among tech giants like Google and Amazon that design their own chips, but it would require substantial investment. Alternatively, OpenAI could acquire a chip company, similar to Amazon’s acquisition of Annapurna Labs in 2015. However, if OpenAI decides to develop its own chips, it would take several years, leaving the company dependent on commercial providers like NVIDIA and AMD in the meantime. The company's motivation to secure more AI chips stems from the shortage of advanced processors and the high costs associated with running the necessary hardware to power its AI technologies.
Key takeaways:
- OpenAI is considering developing its own AI chips or acquiring a chip company due to the scarcity of expensive AI chips it heavily relies on.
- The company is exploring multiple options, including building its own chips, collaborating more closely with chipmakers like NVIDIA, and diversifying its suppliers.
- Developing custom chips would place OpenAI among tech giants like Google and Amazon that design their own chips, but it would require a substantial investment and take several years to develop.
- OpenAI spends up to $700,000 a day maintaining its underlying infrastructure and server costs, and recorded total losses of $540 million in 2022.