Google has invested about $300 million in artificial intelligence startup Anthropic, the latest tech giant to throw money and computing power behind a new generation of companies trying to carve out a place in the booming field of “generative AI.”
Terms of the deal, in which Google will take a stake of about 10 percent, require Anthropic to use the money to buy computing resources from the search company’s cloud computing division, according to three people familiar with the arrangement.
Google’s move highlights the influence some Big Tech companies have over other companies working on AI, which need access to cloud computing platforms to handle giant AI models developed by groups such as Anthropic.
The search firm’s investment also mirrors the $1bn cash-to-compute investment Microsoft made in OpenAI three years ago.
Microsoft’s deal set OpenAI on the path to building a series of breakthrough AI systems, culminating in the release late last year of ChatGPT, a chatbot that can chat with users via text. The software giant followed last month with what it described as a second, “multiyear, multibillion-dollar” investment in the company.
OpenAI and Anthropic are working to develop generative AI, advanced computer programs that can write scripts and create art in seconds.
While Microsoft has sought to integrate OpenAI technology into many of its own services, Google’s relationship with Anthropic is limited to acting as the company’s technology supplier in what has become an AI arms race, according to people familiar with the arrangement.
Anthropic was formed in 2021 when a group of researchers led by Dario Amodei left OpenAI after a disagreement over the direction of the company. He worries that Microsoft’s first investment in OpenAI will set it on a more commercial path and dilute its original focus on advanced AI security.
Anthropic has developed an intelligent chatbot called Claude, competing with OpenAI’s ChatGPT, although it has not yet been publicly released.
The start-up has raised more than $700m ahead of Google’s investment, which was made by the end of 2022 but had not been previously announced. The largest investor in the company is Alameda Research, the crypto hedge fund of FTX founder Sam Bankman-Fried, which put up $500 million before filing for bankruptcy last year. The bankruptcy estate of FTX has flagged Anthropic as an asset that can help creditors with recovery.
Google’s investment was made by its cloud division, run by former Oracle executive Thomas Kurian. Bringing Anthropic’s data-intensive computing work to Google’s data center is part of an effort to capitalize on the leadership Microsoft has enjoyed in the fast-growing AI market thanks to its work with OpenAI. Google’s cloud division is also working with other start-ups such as Cohere and C3 to try to secure a bigger foothold in AI.
Google and Anthropic declined to comment on the investment, or what they said, if any, the tech giant would own in the startup.
According to filings in Delaware, where Anthropic is incorporated, it has two classes of shares, one of which has 10 times the voting rights of the other. Dual-class stock arrangements like this one are often used by tech company founders to ensure they retain control and can dilute the influence of outside investors.
Additional reporting by Tim Bradshaw in London