Amazon's latest investment in Anthropic has cemented the AI startup's position as a major player. Here's what you need to know about the OpenAI rival.
Amazon
Amazon's further investment in Anthropic appears to have cemented the startup's position as a major player in the booming artificial intelligence sector.
Earlier this week, Amazon said it'd commit a further $2.75 billion to the San Francisco-based company as part of its efforts to get ahead of other Big Tech competitors in the AI arms race. It's the e-commerce giant's latest investment into the maker of Claude, a generative AI chatbot that rivals OpenAI's ChatGPT, after pouring $1.25 billion into Anthropic last September.
Amazon had said it'd invest up to $4 billion in Anthropic; this latest outlay brings it to the full amount, showing it's all in on the startup.
The transaction includes a previously announced minority ownership stake. Anthropic also said it will use Amazon Web Service's cloud servers and chips to train and power its large language models, allowing them to produce desirable human-like outputs.
So, what's the deal behind the company Amazon is betting billions on?
Anthropic was founded in 2021 by siblings Dario Amodei and Daniela Amodei, CEO and president of the firm, respectively, after they left their jobs at now-competitor OpenAI to focus on building what they describe as safe AI.
Its mission: "to ensure transformative AI helps people and society flourish," according to the company's website.
During its early days, Anthropic focused on research, training, and testing efforts to ensure its AI model acted in a way that aligned with human values. Then, in March 2023, Anthropic launched the inaugural version of Claude, which could do things like summarize text, write code, and answer questions in a human-like way.
And while OpenAI's ChatGPT came out months before Claude did, Anthropic claimed at the time that Claude was "much less likely to product harmful outputs, easier to converse with, and more steerable." However, some critics said its more cautious approach to an AI chatbot makes its outputs less interesting.
"I would certainly rather Claude be boring than that Claude be dangerous," CEO Amodei told the New York Times last July.
Since the initial launch, Anthropic has released a slate of new AI models. The company announced the Claude 3 family in March this year. That includes Haiku, Sonnet, and Opus, models the startup says produce more accurate outputs and perform more complex tasks than their predecessors.
As of March 28th, Opus, which Anthropic claims is its most intelligent model, appears to perform better than GPT-4, the LLM behind ChatGPT, according to Hugging Face's chatbot arena board, which AI researchers use to gauge the computing systems' capabilities.
Other big players in the tech community have also invested in the company. Anthropic has raised about $9.3 billion as of March 27 — the date of Amazon's further investment announcement — according to PitchBook data. In February, before Amazon's latest cash influx, the company's valuation stood at $15 billion, the NYT reported.
Google pledged to invest up to $2 billion in 2023, and crypto trading platform FTX, under Sam Bankman-Fried's leadership, invested $500 million (though the now-bankrupt FTX is selling a majority of the stake for $884 million to a suite of buyers, CNBC reported earlier this week.)
Moving forward, Anthropic plans to release more updates to its Claude 3 model family and make its AI more suitable for companies.
"We do not believe that model intelligence is anywhere near its limits," Anthropic says.
"Generative AI is poised to be the most transformational technology of our time, and we believe our strategic collaboration with Anthropic will further improve our customers' experiences, and look forward to what's next," Swami Sivasubramanian, vice president of Data and AI at AWS, said in an Amazon press release.
Anthropic didn't immediately respond to a request for comment from Business Insider before publication.
