Amazon invest up to $25 billion in Anthropic part of AI infrastructure

Amazon agreed to invest up to $25 billion At Anthropic, it’s on top of the $8 billion it has poured into its AI startup in recent years as part of an expanded deal to build AI infrastructure.

Inside announcement On Monday, Anthropic said it has committed to spending more than $100 billion on Amazon Web Services technologies over the next 10 years, including current and future generations of Trainium, Amazon’s proprietary AI chips. Anthropic said it has secured up to 5 gigawatts of capacity to train and deploy Claude AI models.

“Anthropic’s commitment to run large language models on AWS Trainium for the next decade reflects the progress we’ve made together on custom silicon as we continue to deliver the technology and infrastructure our customers need to build with generative AI,” Amazon CEO Andy Jassy said in a statement.

Amazon’s investment currently includes $5 billion in Anthropic, with up to $20 billion in the future contingent on “certain business milestones,” according to a statement. The initial investment is at $380 billion, which is Anthropic’s latest valuation.

Anthropic said in a statement that it will bring a total of approximately 1 gigawatt of Trainium2 and Trainium3 capacity online by the end of the year.

While all the major hyperscalers are racing to build AI capacity as quickly as possible, Amazon said in February that it expects to spend about $200 billion on capital spending this year, mostly on AI infrastructure.

Amazon’s investment comes just two months after the e-commerce giant agreed to invest up to $50 billion in Anthropic’s chief rival, OpenAI. Two artificial intelligence companies are racing to convince investors of their strengthening positions ahead of potential IPOs later this year. OpenAI executives have criticized Anthropic in recent months for making a “strategic misstep of not acquiring enough computing.”

Anthropic said in a statement on Monday that the “sharp increase” in consumer usage, as well as enterprise and developer demand for Claude, has led to “inevitable strain” on its infrastructure, affecting its reliability and performance. The company said that its new agreement with Amazon will rapidly increase its existing capacity.

“Our users are telling us that Claude is becoming increasingly important to the way they work, and we need to build the infrastructure to keep up with rapidly growing demand,” Anthropic CEO Dario Amodei said in a statement. he said. “Our collaboration with Amazon will allow us to continue advancing AI research as we deliver Claude to our customers, including more than 100,000 buildings on AWS.”

Anthropic was founded in 2021 by a group of researchers and executives who left OpenAI. The company is best known for its Claude family of AI models and has seen early success selling to businesses. Annual revenue exceeded $30 billion.

Anthropic named AWS as its own primary cloud provider in 2023 and primary training partner however, the company has also signed agreements with rival providers. Microsoft And Google.

In November, Microsoft agreed to invest up to $5 billion in Anthropic, and Anthropic said it had committed to purchasing $30 billion of Azure computing capacity. Earlier this month, Anthropic expanded its partnership with Google and broadcom for “more than one gigawatt” capacity.

— CNBC’s Kate Rooney Contributors to this report

WRISTWATCH: OpenAI memo criticizes Anthropic