Amazon is significantly expanding its stake in Anthropic, the AI laboratory founded by former OpenAI executives, in a deal that could reach $25 billion. The agreement, which includes an immediate $5 billion infusion and up to $20 billion in milestone-based payments, marks a pivotal moment in the industry’s consolidation. It follows $8 billion in previous investments, positioning Amazon as the primary benefactor and infrastructure provider for one of the world’s most capable AI developers.
The partnership is less a traditional venture investment and more a strategic infrastructure swap. In exchange for the capital, Anthropic has committed to spending $100 billion on Amazon Web Services (AWS) over the next decade. Crucially, the lab will continue to utilize Amazon’s custom Trainium silicon, securing up to 5 gigawatts of chip capacity to train and deploy its models. This commitment allows Amazon to validate its own hardware designs against Nvidia’s dominance while ensuring Anthropic has the massive electrical and computational power required for the next generation of large language models.
For AWS customers, the deal simplifies the adoption of Anthropic’s Claude platform by integrating it directly into the AWS portal. By removing the friction of separate credentials and billing, Amazon is effectively folding Anthropic’s "intelligence" into its own cloud ecosystem. This arrangement reflects the emerging reality of the AI era: the most powerful models are increasingly tethered to the massive data centers and proprietary silicon of the world’s largest tech conglomerates.
With reporting from Engadget.
Source · Engadget
