AWS and OpenAI have announced a major expansion of their partnership that brings OpenAI's frontier models, Codex coding agent, and managed AI agents directly into Amazon Bedrock, AWS's managed AI service. The move allows enterprise customers to deploy cutting-edge AI capabilities without leaving their existing cloud infrastructure, security controls, or procurement processes. The new offerings are currently available in limited preview and represent a significant shift in how large organizations can adopt AI development tools at scale.

What Does This Partnership Actually Mean for Enterprise Development Teams?

The expanded collaboration addresses a critical pain point for large organizations: the desire to use advanced AI tools while maintaining control over security, governance, and costs. Rather than forcing companies to adopt separate AI platforms or change their operational frameworks, AWS and OpenAI are embedding these capabilities directly into tools enterprises already use and trust.

The partnership introduces three main offerings. First, OpenAI's latest models, including GPT-5.5, are now accessible through Amazon Bedrock APIs alongside models from other providers like Anthropic, Meta, Mistral, Cohere, and Amazon. This unified access lets enterprises manage model selection, security controls, compliance, and cost governance within a single environment. Second, Codex, OpenAI's AI-powered coding agent, is being integrated into Amazon Bedrock with support for the Codex CLI, desktop application, and Visual Studio Code extension. More than 4 million people use Codex each week for tasks such as writing code, explaining systems, refactoring applications, generating tests, and updating legacy software. Third, Amazon Bedrock Managed Agents powered by OpenAI enables enterprises to build and deploy production-ready AI agents using OpenAI's reasoning capabilities while leveraging AWS's scalable infrastructure and security frameworks.

How Can Organizations Implement These New AI Tools Securely?

• Unified Access Control: All OpenAI models and tools are integrated into AWS security features including encryption, identity and access management, private connectivity, and audit logging, allowing organizations to use OpenAI capabilities without changing existing operational frameworks.
• Bedrock Configuration: Companies with an AWS commitment and Bedrock access can start using Codex by configuring Bedrock as the provider, with all customer data processed through Amazon Bedrock and eligible usage counting toward existing AWS cloud commitments.
• Multi-Step Workflow Support: Amazon Bedrock Managed Agents are designed to support systems that retain context, carry out multi-step workflows, use tools, and take action across business processes, with built-in orchestration, deployment, and governance features.
• Compliance Integration: The managed agent service includes integration into Amazon security and compliance controls, helping organizations move agent projects from early testing into production while maintaining alignment with internal operational standards.

Why Is This Timing Significant for the AI Development Market?

The partnership reflects a broader industry shift toward embedding AI capabilities within existing enterprise cloud environments rather than requiring separate adoption. As organizations evaluate how to deploy generative AI tools, they increasingly prioritize solutions that integrate with their current infrastructure, security protocols, and procurement processes. This move positions AWS as a comprehensive AI platform where enterprises can access multiple leading models and tools without fragmenting their technology stack.

The launch also highlights how coding tools have become a major focus of the enterprise AI market. By linking Codex to Bedrock, OpenAI and AWS are targeting software teams that want AI assistance inside familiar development tools while keeping spending, access, and governance tied to AWS accounts. The managed agents product points to growing demand for AI systems that can carry out sequences of work rather than simply answer prompts, addressing enterprise concerns about oversight, security, and integration with existing business processes.

"The expanded collaboration is designed to give organisations access to advanced AI capabilities while maintaining enterprise-grade security, governance, and operational control," according to AWS's announcement of the partnership.

Amazon Web Services, Partnership Announcement

Enterprise software company Box highlighted the significance of the collaboration, noting that organizations are increasingly seeking to deploy AI agents that can operate at scale while maintaining strong governance, auditability, and security standards. AWS described the launch as the beginning of a deeper collaboration aimed at continuously delivering new AI capabilities through Amazon Bedrock, with future updates expected to bring additional OpenAI advancements in reasoning and agentic systems directly into AWS infrastructure as the technology evolves.

The initial release remains in limited preview, suggesting wider commercial availability will be phased in as both companies test demand and deployment patterns. For organizations already standardized on AWS, this partnership provides a clear path to adopting OpenAI's most advanced models and agents without requiring new infrastructure investments or security reviews.