Amazon Commits $4 Billion Investment into Anthropic, an OpenAI Competitor

Amazon’s recent $4 billion investment in Anthropic, a prominent competitor to OpenAI, marks a significant move in the ever-evolving landscape of artificial intelligence. This strategic partnership will leverage Anthropic’s innovative AI products within Amazon Web Services (AWS), positioning it as a contender against the powerful Microsoft Azure-OpenAI alliance.

Founded in 2021, Anthropic has raised a total of $1.6 billion prior to this latest deal, making it the largest fundraising round to date for the developer of the Claude AI model. While Amazon’s investment sees it take a minority stake, previous funding from companies like Google and Salesforce has been at smaller levels.

The collaboration between Amazon and Anthropic focuses on developing advanced foundation models. Following this partnership, AWS will be the primary cloud provider for Anthropic, although the startup has also established a connection with Google Cloud earlier this year. This multi-cloud strategy reflects a broader trend, as 87% of businesses utilize multiple cloud services to maximize efficiency and flexibility.

As part of the integration, Anthropic is enhancing the capabilities of its Claude foundation model on Amazon Bedrock, a fully managed service that provides clients access to various foundation models via API. Currently, Claude is available on Bedrock, with enterprise users given the ability to customize and fine-tune models. Access to the advanced Claude 2 model is anticipated shortly.

Corporate clients leveraging Anthropic’s models on AWS will be equipped to create generative AI applications designed to automate an array of processes—from generating market forecasts and research reports to improving customer engagement. Dario Amodei, CEO and co-founder of Anthropic, highlighted the substantial organic adoption Claude has received from AWS customers, stating, “By significantly expanding our partnership, we can unlock new possibilities for organizations of all sizes.”

In terms of technical development, Anthropic will utilize Amazon's Trainium and Inferentia chips to train its foundation models, contrasting OpenAI's reliance on Nvidia. Furthermore, Anthropic will have a role in shaping future AWS chip development to better accommodate AI workloads.

This strategic partnership not only enhances AWS’s offerings but also fills a competitive gap in the cloud market, where Microsoft leverages its investment in OpenAI to integrate ChatGPT and other AI services into Azure. As of December 2022, AWS held a commanding 32% market share, while Azure and Google Cloud followed with 23% and 10%, respectively.

A key highlight of Anthropic’s philosophy is its commitment to safer AI outputs, underpinned by its development of ‘constitutional AI.’ This approach involves training models to respond appropriately to challenging questions based on a set of guiding principles, aimed at minimizing harmful outputs.

Anthropic was among the initial cohort of AI companies to endorse the White House’s AI pledge, advocating for increased funding for agencies tasked with establishing standards, including the National Institute of Standards and Technology (NIST). Founded by former OpenAI engineers who sought alternatives after OpenAI's partnership with Microsoft, Anthropic has garnered substantial backing from various investors, including Zoom and Salesforce.

Despite Amazon's substantial investment, Anthropic emphasizes that its governance will remain unchanged. The company operates under the oversight of the Long Term Benefit Trust, which consists of five independent members without financial stakes in the company. This trust aims to ensure that AI model development aligns with what Anthropic terms the “long-term benefit of humanity.” Additionally, as a Delaware Public Benefit Corporation, Anthropic’s board is tasked with balancing corporate interests with public responsibilities, ensuring ethical considerations remain at the forefront of its operations.