OpenAI Investigates Service Disruptions Affecting Codex and Realtime API

OpenAI is currently investigating several service disruptions, according to ChainCatcher. The official status page indicates that the Codex 5.5 engine is experiencing a high error rate, significantly impacting free users. Additionally, all Codex users are facing increased latency with GPT 5.5. The Realtime API's SIP/WebRTC processes have also been interrupted, with the issue marked as 'under investigation' by OpenAI. These disruptions have persisted for several hours. The impact of these issues varies depending on the subscription level and the models and features being used.