Nvidia, OpenAI, Anthropic and Google execs meet with White House to talk AI energy and data centers

3 weeks ago 9

Leaders astatine OpenAI, Anthropic, Nvidia, Microsoft, Google and respective American powerfulness and inferior companies met Thursday astatine the White House to sermon the aboriginal of artificial quality vigor infrastructure successful the U.S., sources acquainted with the gathering told CNBC.

Nvidia CEO Jensen Huang, OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Microsoft president Brad Smith, Google President Ruth Porat and Amazon Web Services CEO Matt Garman each attended the meeting, which focused connected bringing the nationalist and backstage sectors unneurotic to speech astir artificial intelligence's vigor usage, information halfway capacity, semiconductor manufacturing, and grid capacity, sources acquainted with the gathering confirmed.

The White House precocious connected Thursday announced a caller task unit aft the gathering to coordinate argumentation crossed government.

"We're astatine the opening of a caller concern revolution," Nvidia CEO Jensen Huang told CNBC arsenic helium near the meeting. "This manufacture is going to beryllium producing intelligence, and what it takes is energy... So we've got to marque definite that everybody understands the needs coming, the opportunities of it, the challenges of it, and doing it successful the astir businesslike and scalable mode we can."

"There's astir apt going to person to beryllium public-private collaboration successful this area, due to the fact that the complaint of maturation is truly rather high," Huang said, adding that Nvidia is successful the process of starting full-volume accumulation for its Blackwell chip.

An OpenAI spokesperson told CNBC that the institution believes gathering further infrastructure successful the U.S. is captious to the country's concern argumentation and economical future. "We admit the White House convening this gathering arsenic it is simply a designation of the precedence of infrastructure to make jobs, assistance warrant that the benefits of AI are wide distributed, and guarantee America volition proceed to beryllium astatine the forefront of AI innovation," the OpenAI spokesperson said.

OpenAI shared its economical interaction investigation with Biden-Harris medication officials, including the estimated impacts connected jobs and gross home merchandise of gathering a large-scale information halfway successful illustration states crossed the U.S. specified arsenic Wisconsin, California, Texas and Pennsylvania, a root acquainted told CNBC.

CNN archetypal reported connected the meeting.

"President [Joe] Biden and Vice President [Kamala] Harris are committed to deepening U.S. enactment successful A.I. by ensuring data-centers are built successful the United States portion ensuring the exertion is developed responsibly," White House spokesperson Robyn Patterson told CNBC.

Commerce Secretary Gina Raimondo and Energy Secretary Jennifer Granholm besides attended Thursday, according to a root familiar.

The gathering included U.S. National Security Advisor Jake Sullivan; Ali Zaidi, nationalist clime advisor; Kristine Lucius, home argumentation advisor to the vice president; and John Podesta, elder advisor to the president for planetary clime policy. White House main of unit Jeff Zients and White House lawman main of unit Bruce Reed besides attended, according to a source.

The quality follows an announcement successful August that OpenAI and Anthropic volition fto the U.S. AI Safety Institute trial their caller models earlier releasing them to the public, pursuing accrued concerns successful the manufacture astir information and morals successful AI.

The institute, housed wrong the Department of Commerce astatine the National Institute of Standards and Technology, said successful a press release at the clip that it would get "access to large caller models from each institution anterior to and pursuing their nationalist release."

The radical was established aft the Biden-Harris medication issued the U.S. government's first-ever enforcement order on artificial intelligence in October 2023, requiring caller information assessments, equity and civilian rights guidance and probe connected AI's interaction connected the labour market.

OpenAI is reportedly successful talks to rise a backing circular that would worth the institution at more than $150 billion. Anthropic, founded by ex-OpenAI probe executives and employees, was astir precocious valued astatine $18.4 billion. Anthropic counts Amazon as a starring investor, portion OpenAI is heavy backed by Microsoft.

— CNBC's Megan Cassella contributed reporting.

Read Entire Article