Pulsetic can detect an outage before the provider officially reports it.
223 incidents on record since Apr 2026
Last updated: just now · 26 user reports (24h)
CRITICAL · INVESTIGATING
Unable to start new Workspace Agent runs
Started 1 hour ago · Nov 11, 2026 · 10:50 PM UTC
We are investigating an issue preventing users from starting new Workspace Agent runs. We will share another update as more information becomes available.
Having issues too?
User Reports: Last 24 Hours
26 reports
Bars represent 20-min intervals. Occasional spikes may reflect individual issues, not widespread outages.
Outage Map
26 reports (24h)
Reports by region (last 24h)
North America 15(58%)
Europe 5(19%)
Asia 3(12%)
South America 2(8%)
Africa 1(4%)
Click an issue to upvote it and help us better indicate the service status.
Error messages11
Network / connectivity issues6
Timeouts / no response4
Service outage3
Feature not working3
Degraded performance1
Authentication issues0
API errors0
Billing or usage issues0
Other0
Have a different problem?
Issue breakdown
28 total votes
Distribution of reported issues over the last 24 hours.
Error messages39%
Network / connectivity issues21%
Timeouts / no response14%
Service outage11%
Feature not working11%
Degraded performance4%
Authentication issues0%
API errors0%
Billing or usage issues0%
Other0%
90-Day Health Summary
Total Incidents
102
Aug 14 to Nov 12, 2026
Major / Critical
12
12% of the total
Minor Incidents
90
Avg. 1.0 per day
Median Resolution
1h 38m
P95: 20h 37m
Across all 4,387 services Pulsetic tracks, the incidents their status pages reported over the last 12 months had a median resolution time of 1h 58m, and 33% were rated major or critical by the provider. These numbers are calculated separately from the 90-day figures for ChatGPT above.
Check the status badge at the top of this page for the current status. Pulsetic follows the official status page and cross-references it with user reports to give you the most accurate picture.
How often does ChatGPT go down?
Based on our 90-day data, we've recorded 102 incidents, of which 12 were major or critical. The median resolution time is 1h 38m.
How are outages detected?
Pulsetic follows the official status page and records the incidents the provider posts there. Crowdsourced user reports and Pulsetic's own synthetic checks add an early warning, which can surface an issue before the official status page mentions it.
Is it down for everyone or just me?
The outage map and region breakdown help answer this. If reports are concentrated in one area it may be regional; if hotspots appear globally it's a widespread outage. Submit your own report to help others understand the scope.
What does "never acknowledged" mean?
"Never acknowledged" means Pulsetic recorded a service disruption from user reports and its own checks, but the official status page never posted an incident for it. This is one of the core reasons teams use third-party monitoring alongside official status pages.
Get an email when ChatGPT opens, updates or resolves an incident.
Add it as a dependency monitor. The Free plan includes one. See how it works