Amazon is doubling down on its partnership with Anthropic, one of the primary rivals to OpenAI. Amazon’s cloud unit, Amazon Web Services (AWS), today (Nov. 22) announced that it will invest an additional $4 billion into the A.I. startup in an expanded deal that will see Anthropic use AWS chips to train and deploy future A.I. models. The move brings Amazon’s total investments in Anthropic to $8 billion, strengthening the ties between the two companies and deepening their competitive standing against other powerhouse A.I. partnerships like that of Microsoft (MSFT) and OpenAI.
AWS has backed Anthropic since last September when it invested an initial $1.25 billion into the company. Followed by another $2.75 billion this March that represented the largest external investment in Amazon’s history, the partnership positioned AWS as Anthropic’s primary cloud provider.
Under the new investment, Amazon is set to become Anthropic’s “primary training partner,” according to the companies, and will see its Trainium and Inferentia chips used to develop Anthropic’s Claude family of A.I. models. As part of their tie-up, AWS and Anthropic said Claude will continue to be deployed across AWS’s Bedrock platform, where the models are already used by companies like Pfizer (PFE), Intuit, Perplexity AI and the European Union. “We’re looking forward to working with Amazon to train and power our most advanced A.I. models using AWS Trainium, and helping to unlock the full potential of their technology,” said Dario Amodei, Anthropic’s CEO, in a statement.
Anthropic’s primary product is its Claude chatbot, a direct competitor to OpenAI’s ChatGPT. Co-founded by former OpenAI employees in 2020, the San Francisco, Calif.-based company has recently poached a number of top OpenAI leaders and lauds its products as a safety-focused alternative to OpenAI’s GPT. The fast-growing startup is reportedly in talks for a new fundraising round that would value the startup at between $30 billion and $40 billion, doubling the $15 billion valuation it received earlier this year.
Amazon’s ambitious bet on Anthropic parallels a strategy pioneered by Microsoft in 2019, when it first invested in OpenAI. The investment has since grown into a $13 billion partnership. OpenAI’s GPT models are integrated into various Microsoft products, and, in turn, Microsoft provides OpenAI with additional computing power with Azure acting as the A.I. company’s exclusive cloud provider.
The sustainability of such A.I. partnerships, however, is unclear as regulatory scrutiny continues to build around tech leaders. Anthropic and Amazon earlier this year faced a probe from the U.K.’s Competition and Markets Authority (CMA) that examined whether their partnership constituted a de facto merger. While the investigation was later cleared, the regulatory body is currently inquiring into whether Anthropic’s connections to Google (GOOGL), which invested $2 billion into the A.I. startup last year, represents an antitrust threat.
OpenAI and Microsoft, too, have been subject to similar investigations from the CMA and an additional review from the European Commission. Their relationship is also reportedly the subject of an antitrust investigation from the Federal Trade Commission in the U.S., which in January set a broader antitrust examination into motion when it requested information from Microsoft, OpenAI, Amazon, Anthropic and Google on their respective arrangements.
