Notion AI had a 2h 22m incident on 10 November 2026 where some Notion Agent sessions hung or stopped updating. The fault sat on Notion's side, inside the Agent feature, and the only fix on record was refreshing the page.
Started
19:54 UTC
Duration
Lasted 2h 22m
Source
IsDown
Next time
Notion is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: Notion AI
What happened?
Notion logged one incident on 10 November 2026, starting at 19:54 UTC and running for 2h 22m. The affected component was Notion AI, specifically the Agent sessions that handle chat style interactions. Notion's single update, filed at 19:54 as "identified", says some Notion Agent sessions may hang or stop updating, so new responses may not appear in the chat, and that refreshing the page restores the session. That is the only update on record for this incident. No further update was posted to confirm resolution, explain a cause, or say how many sessions were affected. No user reports were filed during the window, so the only account of what happened comes from Notion itself. One filing from a vendor carries only as much weight as its detail, and here the detail is thin: a symptom and a workaround, nothing about scope or root cause. Notion's status history for the past 90 days shows 27 incidents with a median duration of 3 hours 43 minutes, so this one ran shorter than typical. Over all time Notion has logged 330 incidents, averaging 4.3 a month, which puts this kind of disruption in the category of routine for the service rather than exceptional.
Learning
This incident sat inside Notion AI's Agent sessions, a layer that handles chat state on Notion's servers, not something that lives on your own infrastructure or inside requests you control. If you were only watching your own endpoints, you would have seen nothing wrong, because the fault was in how Notion generated and delivered Agent responses, not in your connection to Notion. UptimeRobot's own 30 day figures for Notion show 10 incidents, all resolved, with a median duration of 5h 56m, which is longer than the window this specific incident ran, a reminder that the provider's reported durations and independently observed durations don't always line up. A hung chat session with no new responses is exactly the kind of failure that looks like "my integration broke" from inside your product, when the actual cause is upstream and outside your control. Refreshing the page as a workaround tells you this was a session state problem, not a total outage of the AI feature. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.