OpenAI's Daybreak cyber models arrive on Amazon Bedrock

OpenAI's Daybreak Blue and Daybreak Red cyber capabilities are now available through Amazon Bedrock for eligible customers. Blue provides access to frontier general-purpose models with safeguards for authorized defensive security work; Red is purpose-trained for vulnerability research, exploit validation, and security testing. OpenAI says the models are intended to speed work from vulnerability discovery through mitigation and validation, inside AWS environments where teams already manage security and operations. Access requires enrollment in Daybreak Access; approved customers can use the Amazon Bedrock console or the Responses API through the bedrock-mantle endpoint. 1

The practical takeaway: this is less about adding another chatbot and more about putting specialized cyber models inside an existing enterprise control plane. Teams still need authorization, review, and governance; the announcement does not promise unrestricted offensive use. 1