Daybreak RED is now available to approved customers in AWS GovCloud (US-East), bringing GPT-5.6 Cyber to government security teams through Amazon Bedrock. The expansion is intended to let eligible teams use the model within AWS environments supporting critical missions.
Daybreak Red provides access to OpenAI’s purpose-trained cybersecurity models for authorized vulnerability research, exploit validation and security testing. GPT-5.6 Cyber can help defenders investigate software weaknesses, reproduce vulnerabilities in controlled settings, assess security findings and evaluate potential fixes. Qualified practitioners should review findings and proposed mitigations before putting them into practice.
The service is designed for verified defenders conducting authorized work. Daybreak Red allows greater latitude for these tasks, with fewer refusals, while retaining identity verification, monitoring and access controls. For use through Amazon Bedrock, inference data is not used to train models, and customers are not required to share Daybreak Red data with OpenAI.
Access requires approval from OpenAI for Daybreak Red as well as provisioning by AWS. Approval alone does not automatically enable access to the model on AWS; approved customers must also work with their AWS account team to enable it.
Sources: OpenAI News.