AWS Weekly Update: Anthropic and Meta Deepen AI Collaboration, Lambda Gains S3 Files Support

Introduction: A Week of Strategic AI Moves

After attending the highly energizing Specialist Tech Conference in Seattle, where AWS experts gathered to exchange ideas on Generative AI and Amazon Bedrock, it's clear that community-driven innovation is accelerating. This week's AWS announcements reflect that momentum, with major partnerships and service enhancements that empower builders. Below, we explore the key headlines, from Anthropic's expanded integration with AWS to Meta's adoption of Graviton chips for agentic AI, and a powerful new Lambda capability for file system access.

Strategic AI Partnerships

AWS continues to strengthen its AI ecosystem through deep collaborations with leading AI companies. Two significant agreements were announced this week, both centered around leveraging AWS's custom silicon and services.

Anthropic Partnership: Claude on AWS Trainium and Graviton

AWS and Anthropic have taken their product collaboration to the next level. Anthropic is now training its most advanced foundation models on AWS Trainium and Graviton infrastructure. This involves co-engineering directly at the silicon level with Annapurna Labs, ensuring computational efficiency from the hardware up through the full stack. The result is optimized performance for building and deploying Claude-powered applications.

Additionally, Claude Cowork is now available in Amazon Bedrock. This feature introduces collaborative AI capabilities, allowing enterprise teams to work alongside Claude as a true collaborator rather than a simple tool. With Claude Cowork, you can deploy within your existing Bedrock environment, keeping data secure within AWS while leveraging Claude for team-based AI workflows.

Coming soon: Claude Platform on AWS—a unified developer experience that enables building, deploying, and scaling Claude-powered applications without leaving AWS. For generative AI builders, this marks a significant step forward in what you can achieve with Claude directly through Amazon Bedrock.

Meta Signs Agreement to Power Agentic AI on AWS Graviton

Meta has signed an agreement to deploy AWS Graviton processors at scale. The initiative starts with tens of millions of Graviton cores to power CPU-intensive agentic AI workloads, including real-time reasoning, code generation, search, and multi-step task orchestration. This partnership underscores the growing importance of energy-efficient, high-performance ARM-based chips for AI inference and processing.

New Service Updates

AWS Lambda Now Supports Mounting S3 Buckets as File Systems

AWS Lambda functions can now mount Amazon S3 buckets as file systems using the new S3 Files feature. This enables your functions to perform standard file operations—such as reading, writing, and listing—without downloading data for processing. Built on Amazon EFS, S3 Files combines the simplicity of a file system with the scalability, durability, and cost-effectiveness of S3.

An important advantage: multiple Lambda functions can connect to the same file system simultaneously, sharing data through a common workspace. This is particularly valuable for AI and machine learning workloads where agents need to persist memory and share intermediate results. Developers can now stream large datasets directly from S3, reducing cold start times and simplifying data management.

Conclusion: A Competitive Edge for Builders

This week's updates reflect AWS's commitment to providing infrastructure and services that accelerate AI development. The Anthropic and Meta partnerships expand the available compute and model options, while the Lambda S3 Files enhancement streamlines data handling. For builders, these developments translate into greater flexibility, performance, and cost efficiency. Stay tuned for more innovations as the AWS community continues to push boundaries.