ChatGPT outage: OpenAI investigates degraded access to ChatGPT, Codex and API

A ChatGPT outage struck thousands of users globally on Monday. OpenAI confirmed it was investigating the issue across ChatGPT, Codex, and its API Platform after reports spiked sharply on Downdetector. OpenAI’s status page said impacted users were “unable to access ChatGPT, Codex and API Platform,” citing “degraded performance” affecting login, conversations, voice mode, and image generation. The disruption began around 10:05 AM ET and peaked within about 30 minutes, with the UK seeing over 7,600 reports at peak versus roughly 1,700 in the US. The outage was uneven. Some users faced complete login failures, others could sign in but could not load conversations, and some saw Codex problems while parts of ChatGPT appeared functional—suggesting an infrastructure-layer disruption rather than a single-point failure. Service began partially restoring within roughly an hour, but OpenAI kept the incident under investigation at the time of publication. For traders, this is an indirect risk signal for AI infrastructure uptime: disruptions to major AI platforms can affect downstream business workflows that rely on API access, potentially tightening enterprise confidence and creating short-lived sentiment swings around AI-tech exposure.