← Back to OpenAI's live status
We started measuring OpenAI on August 20, 2026, which is 13 days ago. Everything below covers that period, not the full 90 days a longer-tracked service would show.
The last 24 hours
What happened most recently, on one time axis: our own probe readings by surface, how many people reported a problem, and what OpenAI's own status page declared. Below it, any public chatter we saw about OpenAI and the extra probe it triggered.
The last 24 hours on one axis: what our probes read for each surface we check, how many people reported a problem, and what OpenAI's own status page declared. Three records side by side; none of them changes another.
- ChatGPT Work seeing elevated errors and latency: partial outage, from 15:00 UTC to 20:30 UTC.
Each probe cell is the worst reading across every region in that quarter hour: one region that did not answer colours the cell, never an average. An amber cell is a reading that answered, but at three times or more its 7-day median for three checks running: slow is not down, and we say which. A filtered reading stays neutral, because a refused probe is not an outage. Report bars count unverified reports, not people we verified; they never change our reading. The status-page lane is OpenAI's own declaration, shown as they made it.
Third-party chatter, labeled
Public posts from the last hour that mention OpenAI next to an outage word (“down”, “outage”, “not working”). This is what other people said in public, collected by keyword: not something we measured, and not the user reports above. It never affects the verdict.
- Bluesky posts1 in the last hour (prior day averaged 0.3/hour)
- Confirmation probe15:30 UTC, Asia-Pacific: their own status feed changing prompted an extra probe. Result: OpenAI answered normally.
- Confirmation probe15:29 UTC, Europe: their own status feed changing prompted an extra probe. Result: OpenAI answered normally.
- Confirmation probe15:28 UTC, US-East: their own status feed changing prompted an extra probe. Result: OpenAI answered normally.
- Confirmation probe15:28 UTC, US-West: their own status feed changing prompted an extra probe. Result: OpenAI answered normally.
- Confirmation probe15:18 UTC, Asia-Pacific: a new incident on their status page prompted an extra probe. Result: OpenAI answered normally.
Availability by region
Availability is measured probes that succeeded, divided by every probe we attempted, per region. Not interpolated, not smoothed, and never computed for a region with zero recorded samples.
| Region | Availability | Probes recorded |
|---|---|---|
| US-East | 100.00% | 6240 |
| US-West | 99.98% | 5820 |
| Europe | 99.98% | 4494 |
| Asia-Pacific | 99.79% | 2873 |
Monthly summary
Tracking OpenAI since Aug 20, 2026. In that time our probes detected 1 outage, 11 min in total. Months before that date are not tracked and carry no figure.
| Month | Uptime | Outages | Downtime |
|---|---|---|---|
| Sep 2025 | not tracked | ||
| Oct 2025 | not tracked | ||
| Nov 2025 | not tracked | ||
| Dec 2025 | not tracked | ||
| Jan 2026 | not tracked | ||
| Feb 2026 | not tracked | ||
| Mar 2026 | not tracked | ||
| Apr 2026 | not tracked | ||
| May 2026 | not tracked | ||
| Jun 2026 | not tracked | ||
| Jul 2026 | not tracked | ||
| Aug 2026 from Aug 20, 2026 | 99.93% | 1 | 11 min |
Uptime here is the share of each month in which our probes had no outage open for OpenAI: two consecutive failed readings from a region opens one, the first clean reading closes it. Any detected outage counts, including one seen from a single region, and an outage that crosses midnight on the first of a month is counted in both months with its minutes split. User reports, public posts and OpenAI's own status page contribute nothing to these figures.
Our incident log
Every outage our own probes detected for OpenAI in the last 90 days, or since we started measuring it if that is more recent. This list is probe-only: OpenAI's own declared incidents appear further down this page in their own attributed section, never merged with what we measured here. User reports live on the live page, in their own labeled block.
Their declared record
OpenAI's own incident history, taken from their public status feed. This is their account of what happened, not our measurement, and it never changes the incident log above or the other way around.
Showing the 10 most recent of 28 incidents OpenAI has declared in the history we hold.
These entries are OpenAI's own declarations, taken from their public status feed, shown separately from the outages we detected ourselves. Their page publishes what they choose to publish: an incident missing here means they did not declare one, not that nothing happened, and an incident here is their account, not our measurement. Neither record ever changes the other.
The scorecard
How OpenAI's status page compared with what we measured, for every outage above that falls inside the window we have been capturing their claims.
We started recording OpenAI's own status page on August 20, 2026, 11 days ago. This comparison covers that period only, we do not reconstruct what they said before it, and we compare only inside outages we measured ourselves.
OpenAI's status page reported all systems operational for 11 minutes while we measured OpenAI down in 3 of our regions.
OpenAI decides what their status page covers, and it may report per component while our check reaches one surface from ten regions. We state what each side measured and leave the conclusion to you. If OpenAI believes a figure here is wrong, we publish corrections as prominently as the original.
Full detail on what a probe does and does not prove is in the methodology page.