An issue with an Oracle data center is impacting some parts of the TikTok U.S. user experience. Creators may temporarily experience lags in posting content while Oracle works to resolve the issue. We appreciate your patience and understanding and will keep you updated. https://t.co/ex7S4vM9yU
— TikTok USDS Joint Venture (@tiktokusdsjv) March 3, 2026
US TikTok users are running into issues again today, with the app crashing, videos freezing at zero views, and uploads failing to go through. The culprit? Another Oracle data center outage.
TikTok USDS confirmed on X that a problem at one of Oracle’s data centers is impacting parts of the US user experience, and creators may hit delays posting content while Oracle works on a fix.
Reddit has been flooded with complaints since this morning. Multiple threads show users reporting app crashes during searches, new videos stuck at zero views for hours, and a broken FYP pushing random unrelated content. There’s some good news though.
This is the second time in just over a month. Back in January, an Oracle data center went down after a winter storm knocked out power, causing days of disruption. Users were left with videos frozen in review and zero views, and the FYP took time to recover too, surfacing completely unrelated content for many people long after the outage cleared.
TikTok is not the only one hit today. CapCut, which is under the same TikTok USDS umbrella, is also down, with users reporting login failures, missing Pro features, and broken exports. We covered that here. Since both apps run off the same infrastructure through the US joint venture, one Oracle outage tends to bring them both down at the same time.
TikTok’s American operations are now run through TikTok USDS, a joint venture where Oracle, Silver Lake, and MGX hold an 80% controlling stake, with ByteDance retaining 19.9%. Oracle provides the cloud infrastructure behind the whole US setup, which is why a data center issue hits both apps at once and takes time to fully shake off.
Predictably, some users on Reddit are calling it deliberate, pointing to the timing of global news events and pushing theories about censorship. Oracle has officially acknowledged the infrastructure issue, though, and the January incident had a documented cause tied to a weather-related power failure. The theories are creative, but the explanation here seems straightforwardly technical.
There’s no ETA on full recovery yet. Oracle says it is on it.
Featured image generated with AI

