ChatGPT Down: The June 10th Global Outage Explained

On June 10, 2025, a significant disruption hit users worldwide as the popular AI platform, ChatGPT, experienced a widespread outage. This sudden unavailability of the service, lasting for several hours, caused considerable inconvenience and frustration for millions globally who depend on it daily. The incident underscored the reliance placed on such advanced AI tools and the notable impact when their services become inaccessible. The scale of this event meant the global downtime affected a vast user base across different time zones and professions.

Tracking the June 10th Outage: When and Where it Hit

The technical difficulties on June 10, 2025, began impacting users around 2:36 AM to 3 AM ET. Reports of the issues quickly surfaced, with many turning to independent trackers like Down Detector (suggested link) which showed a sharp increase in problem reports. OpenAI's official status page (suggested link) also acknowledged the incident, confirming elevated error rates and latency affecting services. This period of unavailability wasn't confined to a single region; it impacted users globally, from North America to Australia, highlighting the international scope of the disruption. The data from sources like Down Detector and OpenAI's own site provided a clear timeline and geographic spread of the incident.

Beyond the Downtime: What ChatGPT Users Actually Experienced

During the hours of the outage, users encountered various frustrating ChatGPT issues. Common problems reported included significant latency, failure to generate responses, and elevated error messages. Specific features also seemed affected, with numerous user reports highlighting difficulties with functionalities like Read Aloud problems. Feedback shared on social media platforms, particularly X (formerly Twitter) (suggested link) and Reddit (suggested link), provided a detailed picture of the user experience during the disruption. These reports captured the real-world impact, from minor delays to complete service failure and specific performance issues like Read Aloud problems.

Not Just ChatGPT: Sora AI Video Also Hit by Outage

The technical problems on June 10, 2025, extended beyond just the ChatGPT interface. OpenAI confirmed that other OpenAI services were also affected. Notably, their text-to-video AI tool, Sora, experienced issues concurrently with the main chatbot disruption. This indicated that the cause was likely a more systemic problem impacting multiple parts of OpenAI's infrastructure rather than an isolated issue within ChatGPT itself. The fact that Sora, a different kind of AI tool, was also down underlined the severity and breadth of the technical event.

OpenAI Responds: How the Outage Was Resolved

In response to the extensive reports and user feedback, OpenAI acknowledged the widespread outage affecting their services. They updated their official channels to indicate they were investigating the root cause and implementing mitigation strategies. Their efforts aimed at restoring normal functionality and addressing the performance degradation. After several hours, OpenAI announced that the primary issues had been resolved, and services were gradually returning to normal, effectively bringing ChatGPT back online for the majority of users. This communication and subsequent fix marked the end of the main period of the global downtime.

After the Outage: Are Performance Issues Fixed?

Following the resolution of the main technical outage, discussions among the user community have continued regarding the platform's overall stability. Some users have reported potential lingering performance issues, such as slower response times or difficulty maintaining conversational context over extended interactions. While not a complete downtime, these observations raise user concerns about whether the system has fully stabilized since the major event on June 10, 2025. The community continues to monitor and discuss the platform's behavior after the outage on forums like the OpenAI Community Forum (suggested link).