OpenAI has commenced exploring different chip options besides those from Nvidia for its artificial intelligence workloads. According to a Reuters report, citing eight distinct sources, the creator of ChatGPT has expressed dissatisfaction with some of Nvidia’s recent decisions and is currently reevaluating its hardware acquisition strategy. This shift is driven by the increasing importance of dedicated microchips utilized for AI inference—the phase where the model generates responses to user prompts.
Sources indicate that AI inference is emerging as the next major battleground in the chip market. While Nvidia maintains its supremacy in training large-scale AI models, moves by OpenAI and others to find substitutes represent a significant threat to their stronghold. This situation unfolds amid ongoing discussions about potential investments and shifting priorities within the broader AI ecosystem.
Grok Image
Previously, Nvidia CEO Jensen Huang had mentioned the company’s intention to be involved in financing OpenAI, deeming such an outlay to be profitable, although specific figures were not disclosed. Furthermore, there had been earlier discussions regarding a potential investment reaching up to $100 billion; however, The Wall Street Journal reported that these plans were put on hold at the end of January due to internal reservations at Nvidia concerning the terms of the arrangement.
About the Author
