ChatGPT suffered a global outage peaking ~12:44 p.m. IST, lasting ~three hours
Downdetector logged ~2,000 US reports and 500–550 from India
Both web and mobile clients were affected, indicating server-side infrastructure issues
OpenAI confirmed restoration but offered no detailed post-mortem; enterprises demand reliability
OpenAI’s ChatGPT experienced a widespread outage on Wednesday that left users around the world unable to access the chatbot for several hours before service was largely restored, according to user reports and outage trackers.
The disruption peaked around 12:44 p.m. IST and lasted roughly three hours at its worst, during which many users could not load the web client or receive replies from the mobile apps.
Outage-monitoring site Downdetector showed hundreds of reports from India (between 500 and 550) and the US (more than 2,000) by mid-afternoon IST.
OpenAI’s public status page acknowledged the incident and later indicated the service had been fixed, but the company has not published a detailed explanation or post-mortem of the root cause.
Scope & User Impact
Reports came from multiple regions including India, the US, the UK and the EU, with complaints describing stuck loading screens, error messages and the inability to submit prompts or retrieve chat histories.
Both the browser experience and native iOS and Android apps were affected, suggesting a server-side problem in OpenAI’s infrastructure rather than a platform-specific bug. While the platform returned to service for most users, some continued to report intermittent problems after the initial fix.
This outage follows several interruptions earlier in 2025: notable incidents in July and June that affected thousands of users, and a January outage that also disrupted service. The recurrence of such incidents has raised concerns among enterprise customers and heavy users who depend on the tool for coding, research, content production and customer support.
OpenAI Response
OpenAI’s status page confirmed the outage and later the restoration of service but has not provided a technical breakdown or timeline explaining what failed and why. Enterprises and developers who rely on the API have been pressing for more detailed communications and assurances about uptime and contingency plans.
As generative-AI tools such as ChatGPT become woven into workflows across education, software development, journalism and business operations, even short interruptions can cause outsized disruption. Recurring outages underscore the challenges of operating large-scale AI services and are likely to intensify calls from customers and regulators for clearer reliability standards and post-incident transparency.
What Users Can Do Now?
If you still face issues, check OpenAI’s status page for live updates, retry after clearing caches or logging out and in, and consider local fallbacks (cached responses, alternative tools or rate-limited API usage) for critical workflows until full stability is confirmed.
The outage highlighted the growing dependency on generative-AI platforms and the operational strain that comes with serving millions of users. The industry will be watching closely for a full explanation from OpenAI and any steps the company takes to bolster resilience and reassure large-scale users.
