OpenAI Models and Codex Launch on AWS
June 1, 2026 — OpenAI’s frontier models and Codex are now generally available on AWS, enabling enterprises to deploy these capabilities within their existing AWS environments.
Availability on Amazon Bedrock
The offerings can be accessed through Amazon Bedrock. One option is OpenAI models on Amazon Bedrock, which integrates OpenAI capabilities with AWS-native security and governance controls. The other option is Codex on Amazon Bedrock; Codex serves as OpenAI’s software engineering agent, utilized by over 5 million users weekly for tasks such as writing, reviewing, debugging, and modernizing code. These services are available in both Commercial and GovCloud regions, allowing organizations to adopt the models while leveraging their existing procurement, billing, security, and governance workflows.
Enterprise Impact and Next Steps
This integration aims to reduce operational friction related to procurement, security review, and production readiness, facilitating a quicker transition from evaluation to deployment. Sean Bruich, Senior Vice President and Chief Technology Officer at Amgen, highlighted that GPT‑5.5 and frontier models provide significant advances in capability, quality, and consistency. He noted that making these models available on AWS offers a pathway to scale within responsible AI frameworks that encompass security and governance.
Ritesh Bansal, VP of Analytics Data, Agentic AI and AI/ML Platform at Autodesk, mentioned that teams are currently evaluating how frontier AI capabilities and AI-powered development tools on AWS infrastructure can enhance development workflows and support informed decision-making for customers.
Future Developments
OpenAI has indicated plans to expand the capabilities available through AWS. This includes Daybreak, an initiative aimed at transforming software development and defense. Daybreak will feature cyber models and Codex Security, designed to assist cyber defenders with secure code review, threat modeling, patch validation, dependency risk analysis, detection, and remediation guidance.
Original source: OpenAI News