OpenAI CEO Sam Altman’s latest stop on his AI Emperor Roadshow was in the United Arab Emirates, where he envisioned a global consortium of governments and private interests to fund, power and deliver the artificial intelligence industry. Presented.
Altman met with Emirati officials and investors this week to discuss ways he and others could work together to offset the steep costs associated with AI infrastructure, according to a report. . Energy and semiconductor supplies, as well as data center capacity, are said to be on the agenda. Altman is said to have also met US Ambassador to the Emirates (UAE) Martina Strong during his visit.
Over the past few months, Altman has apparently solicited billions of dollars from industry giants — including Abu Dhabi-based G42, Japan’s SoftBank, Microsoft, and others — to build a network of chip fabs dedicated to making silicon for machine learning. could
Now it seems the conversation wasn’t just about increasing the supply of AI accelerators that power OpenAI and other models. Rapidly increasing data center capacity and alternative energy sources – including nuclear – were also on the agenda. The latter is no surprise – Altman is a big fan of nuclear power, having previously thrown his weight behind small modular reactor startup Oklo and fusion hopeful Helion Energy.
In a statement to Bloomberg, OpenAI would only reveal that “there are ongoing discussions about expanding the global infrastructure and supply chain for chips, energy and data centers … We will share more details at a later date.” Looking forward to it.”
The UAE — which, needless to say, doesn’t have a great record on human rights — is particularly interested in investing in AI, and has plenty of cash to spread. . The country’s G42 investment group is investing hundreds of millions of dollars in AI infrastructure purchases – including a $900 million plan to build a cluster of supercomputers using Cerebra’s wafer-scale accelerator.
The G42’s affairs, however, have come under scrutiny in recent months amid accusations that China has provided advanced AI technologies and genetic data that describe millions of people. Perhaps to appease the US, the G42 has since cut ties with Chinese suppliers.
Apart from the UAE, Altman has reportedly held similar talks with Western countries. According to Bloomberg, the next leg of Altman’s odyssey will take him back to Washington.
The scope of Altman’s ambitions is unclear at this time. Various anonymous sources have estimated that his proposals could cost billions – or trillions – of dollars. One report claimed that Altman was seeking upwards of $7 trillion for Fabs — a figure we noted at the time is 14 times the total revenue of the entire semiconductor industry in 2023.
Altman has since denied that he is trying to raise trillions of dollars. Speaking to Intel CEO Pat Gelsinger during the x86 giant’s Foundry Direct Connect event in February, the OpenAI boss reminded everyone that not everything you read on the Internet is true.
While Altman may not need $7 trillion, he argued that advancing AI will likely cost more than people think, and will require investment on a global scale. “I think everyone is underestimating the need for a lot of AI computing,” the AI supremo told Gelsinger.
One thing is certain: the power-hungry GPUs and high-speed AI used to train and run models remain in short supply despite efforts to increase output. Even with Nvidia — a leading producer of AI infrastructure — expecting to more than triple production of its H100 and H200 accelerators this year alone, analysts warn that demand will outstrip supply in the near future. will ®