“Microsoft has successfully deployed OpenAI's GPT models on AMD hardware for inference workloads.”