Nvidia is all in on ChatGPT.
During the company’s GTC 2023 keynote address on Tuesday, March 21, 2023, Nvidia CEO Jensen Huang highlighted the company’s accelerating focus on AI, which Nvidia hopes to bring to every industry possible.
“The warp drive engine is accelerated computing, and the energy source is AI,” Huang said during the keynote (opens in new tab). “The impressive capabilities of generative AI have created a sense of urgency for companies to reimagine their products and business models.”
The 78-minute long keynote was heavily enterprise focused, with almost all of the event devoted to talking about cloud computing and server based technology like the new Grace server-grade CPU. But there was some talk of more consumer focused products like ChatGPT, the viral large language model that everyone has been using to do everything from write spec scripts to homework, or even pass law exams.
ChatGPT is especially important for AI, which Nvidia is heavily invested in, since its introduction is really an inflection point for AI going mainstream. “We are at the iPhone moment of AI,” Huang said.
Nvidia’s position in the AI marketplace
Nvidia’s focus on AI and ChatGPT might be a surprise for many, since Nvidia isn’t putting out any large language models itself, but Nvidia technology ultimately underpins all of them.
Nvidia’s GPU architecture features advanced tensor cores that are essential to carrying out machine learning data processing and the adversarial generative techniques used to produce images through Stable Diffusion and text through ChatGPT.
Given the expected surge in commercial use of this technology in the coming years, Nvidia is smart to place itself at the center of it all, as it is really the only processor maker with tensor core technology essential to the task of generative AI to the point of having an effective monopoly on the tech.
What this means for other areas of Nvidia’s business, like making the best graphics cards in the world for gamers, remains to be seen, but given the market pressures for AI, it’s not out of the question to see Nvidia slowly pull itself out of the consumer market almost entirely and focus on server-level hardware while leaving consumer GPUs to AMD and Intel.
This isn’t great news for gamers, obviously, since one less choice for GPUs is always a downgrade, but Nvidia hasn’t said yet what its plans are for the future, so this all remains to be seen.
