Artificial Intelligence

Microsoft and Nvidia invest 15 billion in OpenAI challenger Anthropic, strengthening control over AI infrastructure

The two companies focus on Anthropic and turn yet another spotlight on circular activities around AI

by Biagio Simonetta

2' min read

Translated by AI
Versione italiana

2' min read

Translated by AI
Versione italiana

Microsoft and Nvidia (which is due to release its quarterly accounts in the coming hours) have reportedly committed to invest up to USD 15 billion in total in the next round of financing for Anthropic. This was reported by the Bloomberg news agency. And it is a passage that says a lot about where the race for artificial intelligence in the US is heading. That is, towards ever closer concentrations between those who build the models and those who own the infrastructure that runs them. A pattern already seen, moreover, with OpenAI.

The arrangement is part of a now recognisable trend: large cloud and chip providers finance model developers, and the latter, in return, buy computing power from the same entities that finance them. A closed-loop mechanism that is creating more than one alarm among investors, who are convinced that these 'circular AI deals' are the clearest sign of a possible bubble.

Loading...

Inside the deal is a crucial point: Anthropic (which is considered by many to be OpenAI's real rival) commits to buying $30 billion of computing capacity from Microsoft's Azure cloud. It means that the company led by Satya Nadella, already a key investor in OpenAI, continues to strengthen its network of influence in the market for generative models.

Nadella himself makes this clear in a video released in recent hours: 'We will become each other's customers: we will use Anthropic's models, they will use our infrastructure, and we will go to market together. All this is based on the partnership with OpenAI, which remains a key partner for us'. A sentence that also captures the evolution of the last few months, marked by an increasing competition between Microsoft and OpenAI, even within the same industrial alliance.

For Anthropic - founded in 2021 on the initiative of former OpenAI employees and positioned as a company concerned with the security and reliability of AI - the move is twofold: on the one hand it consolidates strategic relations with one of the cloud giants, and on the other it increases its computing power with a view to expanding its models.

The company, which raised $13 billion at a valuation of $183 billion in September and claims to have 300,000 corporate customers, is building a gigantic infrastructure: $50 billion to build dedicated AI data centres in multiple locations in the US, including Texas and New York.

It should be remembered that Microsoft is not the only giant in orbit Anthropic. In October, the company struck a deal with Google, which will supply up to a million specialised AI chips, a tens of billions of dollars contract. And then there is Amazon, a structural partner that has invested $8 billion in the company and built a complex of data centres and AI chips designed by AWS for Anthropic itself.

Copyright reserved ©
Loading...

Brand connect

Loading...

Newsletter

Notizie e approfondimenti sugli avvenimenti politici, economici e finanziari.

Iscriviti