Anthropic seeks huge investment from Google just days after Amazon invested billions
Anthropic has emerged as one of the best-funded rivals to OpenAI LP in generative artificial intelligence, but despite raising billions of dollars already, it’s still seeking more.
According to a report today in The Information, Anthropic is now holding talks with investors, including Google LLC, about a new $2 billion funding round. It comes after Amazon.com Inc. last week committed to invest up to $4 billion into the startup.
The comapny has reportedly told investors that it’s seeking to reach a valuation of between $20 billion and $30 billion following the conclusion of the proposed round. Such a figure is far in excess of the $4 billion valuation that was slapped on the startup back in March, following a previous round. Notably, it would mean Anthropic’s shares are more expensive relative to revenue than OpenAI’s, The Information added.
Anthropic is a two-year-old startup that’s building generative AI models in competition with OpenAI. Its flagship model is Claude 2, which is a chatbot with similar capabilities to ChatGPT, capable of generating marketing copy, solving math problems and holding engaging conversations with humans.
However, Anthropic reportedly has ambitions to build a much more powerful model known as Claude Next, which is so resource-intensive that not even the world’s most advanced supercomputers would be able to power it. As such, it needs to raise an awful lot of capital to proceed with its development.
The startup has mentioned the huge costs of running its AI models as one of the reasons for its decision to launch a paid subscription service called Claude Pro, which allows users to send many more messages and receive priority during busy periods.
The costs of AI are no joke. Earlier this year, OpenAI was reported to be spending $21 million a month on the cloud-based graphics processing units and other computing resources necessary to power ChatGPT.
With Anthropic’s ambitions to build an even more powerful generative AI model, it needs funding and it has sought money from numerous investors this year. In addition to Amazon, it has also secured the backing of SK Telecom Co. Ltd., which invested $100 million in the startup in August. In March, Anthropic closed on a bumper $450 million round that saw participation from Google plus Salesforce Ventures and Zoom Ventures.
Leaked documents earlier this year showed Anthropic is seeking to raise as much as $5 billion over the next two years to build Claude Next. The documents suggest the new model will be 10 times more powerful than any existing generative AI system available today. To build such a system, Anthropic believes it will need access to computing infrastructure that can deliver 10 septillion floating point operations per second, or FLOPs, of performance. That’s a one followed by 25 zeros. It’s several orders of magnitude larger than a quintillion FLOPs, which is the average performance of today’s biggest supercomputers.
That explains why Anthropic needs the backing of cloud computing giants such as Amazon and Google. Amazon, which has an option to invest $2.75 billion more in Anthropic in addition to its initial $1.25 billion, has also agreed to provide a good portion of the cloud infrastructure the startup will eventually need. Amazon Web Services was named as the startup’s “primary cloud provider for mission critical workloads.” It will provide it with its most powerful compute instances, including its new AWS Trainium and AWS Inferentia chips, which are specifically designed for AI training and inference workloads.
Anthropic has also committed to making its future models available via Amazon’s AWS Bedrock service. It will also help the cloud provider refine the capabilities of its dedicated AI chips.
Anthropic’s partnership with Amazon shows that AI startups and cloud infrastructure vendors can both benefit from closer cooperation with one another, said Holger Mueller of Constellation Research Inc. “The large language model creators need enormous computing resources, while the cloud providers want the AI workloads to scale their business and also model diversity, so customers don’t get worried about lock-in,” Mueller said. “It is along these lines that Google is apparently holding talks with Anthropic, which is looking more and more like it will become the funding queen of AI in 2023.”
Nothing has been revealed publicly so far, but it seems likely that Google will want a similar level of collaboration with Anthropic if it’s to participate in what looks to be another enormous round of funding.
Image: Anthropic
A message from John Furrier, co-founder of SiliconANGLE:
Your vote of support is important to us and it helps us keep the content FREE.
One click below supports our mission to provide free, deep, and relevant content.
Join our community on YouTube
Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.
THANK YOU