Google’s Billion-Dollar Push in AI: Investment in Anthropic Expands

Google LLC has reportedly invested over $1 billion in Anthropic PBC, adding to the $2 billion it had already allocated to the artificial intelligence firm. This significant funding, as revealed by the Financial Times, comes alongside Anthropic’s ongoing effort to raise $2 billion in a separate round led by Lightspeed Venture Partners, which is expected to value the company at $60 billion.

Google’s Billion-Dollar Push in AI: Investment in Anthropic Expands
Google’s Billion-Dollar Push in AI: Investment in Anthropic Expands

Anthropic specializes in developing enterprise-grade large language models (LLMs) branded as Claude. These models are capable of generating text, writing code, and performing various tasks, making them versatile tools for businesses. The company provides access to its models through an application programming interface (API), enabling integration into third-party software solutions. In June, Anthropic unveiled its most advanced model yet, Claude 3.5 Sonnet, which the company asserts outperforms OpenAI’s GPT-4o in certain reasoning tasks. Additionally, this model is designed to perform business-related tasks on behalf of users, showcasing its potential in enterprise environments.

At a Wall Street Journal event earlier this week, Anthropic CEO Dario Amodei shared updates on the company’s plans for its AI models. He revealed that Anthropic is working to enhance its LLMs with web-browsing capabilities and the ability to remember user preferences across multiple projects. Furthermore, the company aims to increase the rate limits of its API to accommodate higher volumes of requests from developers. According to Amodei, the current limitations are partially due to the backend infrastructure struggling to meet the growing demand, an issue the company is addressing by upgrading its systems. The reported $1 billion investment from Google is expected to support these efforts by enabling Anthropic to acquire additional AI hardware.

In November, Anthropic secured $4 billion in funding from Amazon.com Inc. as part of a partnership to utilize AWS Trainium chips for training its next-generation foundation models. However, it remains unclear whether Google’s latest investment might lead to Anthropic increasing its reliance on Google Cloud infrastructure, creating potential competition with AWS.

Last month, AWS announced a joint project with Anthropic to develop Project Rainier, an AI supercomputer designed to include hundreds of thousands of Trainium chips. Once operational, this system is expected to deliver five times the computing capacity of the cluster currently used to train Anthropic’s flagship models.

In addition to infrastructure improvements, Anthropic is intensifying its competition with OpenAI by developing more sophisticated LLMs. At the same Wall Street Journal event, Amodei stated that the company is working on models capable of handling more complex tasks. While not exclusively focused on reasoning, these upcoming general-purpose models will include enhanced problem-solving abilities, aligning with Anthropic’s broader vision of advancing AI capabilities.