Artificial intelligence startup Anthropic has committed to spending more than $100 billion on Amazon’s AWS cloud platform over the next decade to train and operate its Claude chatbot. Amazon will invest $5 billion immediately and may contribute up to another $20 billion in the future, the companies announced April 21.
The deal represents a major infrastructure bet as technology companies race to secure computational resources for large language models. The partnership grants Anthropic access to up to 5 gigawatts of Amazon’s custom Trainium chips and deepens integration of Claude into the AWS platform, where 100,000 customers already run the chatbot.
Anthropic and Amazon have expanded their partnership in a major infrastructure deal reflecting the intense competition for computational resources among artificial intelligence developers. The AWS cloud platform will provide the AI startup with custom chips optimized for training and deploying Claude, the chatbot Anthropic has positioned as a rival to OpenAI’s ChatGPT.
The companies have worked together since 2023 to expand AI adoption among AWS customers. The new commitment deepens that relationship and represents Amazon’s confidence in Anthropic as a long-term technology partner despite regulatory headwinds the startup has faced.
Anthropic remains unprofitable but is growing rapidly. The startup reported in February that it expects $14 billion in sales over the coming year, a sharp increase from its operations at the time of its previous funding round. Anthropic was founded in 2021 by former OpenAI employees and released Claude in 2023, months after ChatGPT’s public debut.
The company’s $380 billion valuation, announced earlier this year, positions it alongside other technology giants. Renaissance Capital, which tracks the most promising private companies for potential public offerings, ranks Anthropic third in value behind SpaceX and OpenAI, valued at $500 billion.
The partnership announcement comes as the Trump administration has pursued restrictions on Anthropic. In February, the White House ordered all federal agencies to cease using the company’s AI technology, citing what officials described as national security concerns. President Trump, Defense Secretary Pete Hegseth, and others publicly criticized Anthropic for refusing to give the military unrestricted access to its systems.
Dario Amodei, Anthropic’s CEO, declined to comply with the order. The company said it would contest the action as unprecedented and legally unsound, arguing that its safeguards are essential. A federal appeals court refused to block the Pentagon from blacklisting the company, allowing the restrictions to remain in effect.
The infrastructure investment announced with Amazon underscores that despite government pressure, Anthropic has secured backing from a technology leader with deep resources and established cloud infrastructure. AWS customers, numbering 100,000 who currently run Claude models, will gain direct access to the chatbot through the cloud platform as part of the integrated deployment.