Google’s parent company, Alphabet, has significantly upped its investment in Anthropic. After an initial investment of $300 million in April, the tech giant is now funneling an additional $2 billion into Anthropic.
The payment structure is intriguing: $500 million upfront, with the remaining $1.5 billion to be disbursed over time. This investment secures Google a comprehensive cloud partnership, adding another feather to its already well-plumed cap.
Anthropic’s chatbot, Claude 2, is a formidable rival to OpenAI’s ChatGPT. With an ability to process around 75,000 words in context — equivalent to the length of a book — Claude 2 outperforms ChatGPT, which has a limit of approximately 3,000 words. Claude 2 also scores above the 90th percentile on GRE reading and writing exams, as well as on quantitative reasoning tests.
Google is not alone in recognizing Anthropic’s potential. In September, Amazon announced a partnership with the AI startup, valued at a staggering $4 billion, taking a minority ownership position. Anthropic is also using Amazon Web Services (AWS) Trainium and Inferentia chips for its future foundation models. Other investors in the mix include Salesforce and Zoom, indicating a broader industry trend of big tech companies diversifying their AI portfolios.
Anthropic was founded by siblings Dario and Daniela Amodei, former OpenAI developers. They left OpenAI over disagreements about the ethical direction of AI development. Anthropic is one of the seven companies invited by the U.S. administration to discuss responsible AI development. The company is committed to watermarking AI-generated content for easy identification.
This investment is part of a larger trend among cloud operators who are keen on diversifying their AI models. Microsoft, for instance, has invested over $13 billion in OpenAI since 2019 and has also announced a $3.2 billion partnership with the Australian government. Google itself launched its own ChatGPT competitor, Google Bard, in February and has also initiated a $20 million Digital Futures Project to support responsible AI development.