Home / News / Technology / Anthropic’s Claude 3 Beats OpenAI’s GPT-4 As Rising AI Company Receives $4Bn Amazon Investment 
3 min read

Anthropic’s Claude 3 Beats OpenAI’s GPT-4 As Rising AI Company Receives $4Bn Amazon Investment 

Last Updated March 28, 2024 7:24 PM
James Morales
Last Updated March 28, 2024 7:24 PM

Key Takeaways

  • Claude 3 Opus has overtaken GPT-4 as the leading foundation model on Chatbot Arena.
  • The development ends GPT-4’s prolonged spell at the top of the AI leaderboard.
  • Meanwhile, Amazon will invest a further $2.75 million in Claude’s developer Anthropic.

Since it was first listed on Chatbot Arena  in 2023, different versions of OpenAI’s GPT-4 have held the top spot on the AI leaderboard. But on Tuesday, March 26, the large language model (LLM) was toppled by Anthropic’s Claude 3 Opus.

The development occurred just as Amazon has committed a further $2.75 billion to the Anthropic, bringing the total value of its investment in the AI startup to $4 billion.

Amazon Doubles Down on Anthropic Investment

Amazon initially invested $1.25 billion in Anthropic last year in a deal that gave it the right to increase the value of its investment to $4 billion before the end of March. 

With days to spare, the Financial Times has reported that Amazon will now take up the opportunity to more than double the size of its stake.

On top of the new Amazon money, Anthropic is also raising funds from other investors that could value the company at more than $18 billion, the report claimed.

Amazon’s partnership with Anthropic has seen Claude integrated into various aspects of the firm’s business.

Available to the public via Amazon Bedrock, the model also directly competes for enterprise users with GPT-4, which is offered on Microsoft’s Azure AI platform.

Chatbot Market Has a New Top Dog

Since launching ChatGPT in 2022, OpenAI has dominated the LLM service market.

While the company had a first-mover advantage, the objective superiority of its models also helped it fend of competition from Anthropic, Google and others.

In recent months, however, both Claude 3 and Google’s Gemini have boasted higher scores on AI benchmarks than GPT-4. Although higher test scores alone don’t make those models better per se, the recent Chatbot Arena overtake suggests that Claude 3 has successfully won over users.

Top Foundation Models Neck and Neck

An open-source leaderboard that lets users vote on which AI performs best, Chatbot Arena uses the Elo rating system to rank different models.

Not so long ago, GPT-4 enjoyed a significant lead over its closest rivals on Chatbot Arena. But now, just a few points separates the front of the field.

At the time of writing, Claude 3 had an Elo score of 1253. Meanwhile, GPT-4 Turba scored 1251. Google’s Gemini Pro came in third with 1248.

Was this Article helpful? Yes No