Sam Altman, CEO of OpenAI, who visited Korea to expand the artificial intelligence semiconductor ecosystem and met with top executives of Samsung Electronics and SK hynix, is now set to visit American semiconductor firm Intel.
According to industry sources on Feb. 1, Intel CEO Pat Gelsinger announced on social media platform X (formerly Twitter) on Jan. 30 (local time) that "Altman will meet me on Feb. 21st at Direct Connect."
Gelsinger, who took office in February 2021, launched the Intel Foundry Services (IFS) division to handle foundry operations with a pledge to “restore past glory.” Direct Connect is the first external event hosted by Intel IFS.
Altman is planning to develop his own AI semiconductors to reduce reliance on Nvidia, which currently dominates the global AI semiconductor market.
OpenAI released the latest version of its Large Language Model (LLM), GPT-4 Turbo, at the end of last year and is currently working on further upgrades. For this, a significant amount of the latest AI semiconductors is crucial.
Currently, Nvidia virtually monopolizes the AI semiconductor market with a market share of 80%. This has led to a shortage of AI semiconductors, prompting companies like Google, Amazon Web Services (AWS), and Meta to actively pursue the development of their own AI semiconductors. OpenAI appears to be firmly moving in the direction of producing AI semiconductors in-house as well.
To this end, Altman is making diverse efforts, including contacting semiconductor manufacturers such as TSMC, UAE AI company G42, and the SoftBank Group of Japan, which owns British semiconductor design company ARM. It is also reported that he visited the U.S. Congress to discuss the establishment of semiconductor manufacturing plants.