OpenAI ends exclusive cooperation with Microsoft and fully launches on AWS Bedrock: AI models enter the multi-cloud era
OpenAI and AWS announce an expanded collaboration to bring cutting-edge models like the GPT family and Codex to Amazon Bedrock. This marks a major transition in AI infrastructure from Azure exclusive to multi-cloud coexistence. Developers can directly call OpenAI models to build Agents and automated workflows in the AWS ecosystem.
Core conclusion
On April 29, 2026, OpenAI officially announced an in-depth cooperation with Amazon AWS to introduce its latest AI models (including GPT series, Codex, etc.) to the Amazon Bedrock platform. This is the first major infrastructure collaboration since OpenAI ended its cloud-exclusive partnership with Microsoft, marking the entry into the multi-cloud era of AI model distribution.
Key Points
- Time of incident: 2026-04-29 -Affected objects: All developers using OpenAI API and AWS ecosystem users
- Core change: OpenAI models can be called directly through Amazon Bedrock and are no longer bound to Azure
- Strategic significance: AI infrastructure moves from "single cloud exclusive" to "multi-cloud coexistence"
Background and trigger events
On April 28, OpenAI and Microsoft announced an adjustment in their partnership—OpenAI will no longer maintain exclusive cloud infrastructure binding with Azure. Just one day later, on April 29, OpenAI announced an expanded cooperation with AWS to bring its cutting-edge models (including GPT series, Codex, etc.) to Amazon Bedrock.
According to Stratechery analyst Ben Thompson’s joint interview with OpenAI CEO Sam Altman and AWS CEO Matt Garman, OpenAI’s strategic focus is tilting towards AWS, especially the deep integration with the new capabilities of Amazon Bedrock Managed Agents. Altman said in the interview that AWS's enterprise coverage and infrastructure scale form a "natural combination" with OpenAI's model capabilities.
Source: The Verge, Stratechery, Amazon News Blog
Key Impact (by Dimension)
| Dimensions | Change | What it means to us | Recommended actions |
|---|---|---|---|
| Deployment flexibility | OpenAI models can be called from Azure / AWS / own API simultaneously | No longer locked into a single cloud, enterprise users can choose the optimal infrastructure | Evaluate OpenAI model pricing and latency performance on AWS Bedrock |
| Codex availability | OpenAI Codex (code generation model) comes to AWS | Building AI coding agents in the AWS ecosystem becomes more straightforward | Testing Codex’s agent workflow integration on Bedrock |
| Cost structure | Multi-cloud competition may lead to price optimization | Price competition between AWS and Azure may reduce inference costs | Continuously monitor the pricing differences between the two places and switch as needed |
| Agent Development | Bedrock Managed Agents natively supports OpenAI models | Building complex AI Agents no longer requires cross-platform calls | Try out the combined solution of Bedrock Managed Agents + OpenAI |
| Enterprise Adoption | AWS enterprise customers can invoke OpenAI directly on existing infrastructure | Easier approval for use by enterprise compliance and security teams | Recommend OpenAI models on AWS to customers, reducing compliance friction |
Adaptation suggestions
This change has direct benefits for content automation, AI Agent development and SaaS construction:
- Multi-cloud deployment strategy: No longer dependent on a single cloud provider, you can flexibly switch between Azure and AWS based on performance and cost
- Agent development simplified: Bedrock Managed Agents natively integrates the OpenAI model, reducing the complexity of manually managing API keys and cross-platform calls in Agent logic.
- Enterprise market expansion: AWS has the largest enterprise customer base, and the arrival of OpenAI models in Bedrock means that enterprise users can use cutting-edge AI capabilities while being technically compliant.
- Codex new scenario: Codex is available on AWS, further lowering the threshold for building "AI programming assistant" products
Task List
- Evaluate Bedrock region availability for current AWS account
- Comparing latency and pricing for Azure OpenAI and Bedrock OpenAI
- Test the code generation capabilities of Codex in Bedrock Agent
- If using automation tools such as n8n, configure multi-platform AI calling nodes
Example: Call OpenAI model on Bedrock
# AWS CLI Bedrock OpenAI
aws bedrock-runtime invoke-model --model-id openai.gpt-4o --body '{"messages":[{"role":"user","content":"AI Agent"}]}' --region us-east-1 output.json
cat output.json | jq '.'（）
，OpenAI、ChatGPT、Claude、n8n、LangGraph、Hermes Agent、DeepSeek. For content automation and AI agent developers, new cloud options mean greater flexibility and lower costs.
Internal link guidance
- Want to learn how to use n8n + OpenAI to build automated workflows? See: n8n + OpenAI ：30
- Real case: A developer used Claude Code to start a business within 48 hours: Claude Code 48 hours to start a business: one person + US$29 monthly fee, monthly income in 3 months $9,000
- Use AI to build an automated content publishing system, please refer to: How to build an AI content automated distribution system with n8n + ChatGPT: a complete 30-minute tutorial
Monetization angle
How can you make money from this trend?
WayToClawEarn focuses on verified earn playbooks—not just news. Start from these cases.
n8n + OpenAI affiliate site
Automate content and affiliate monetization
Claude + n8n automation agency
Charge monthly for agent workflow builds