We’re likely getting hit by another wave of web scrapers for AI training, which is causing issues on the back end of the GitLab UI.
This is well known: we aggressively rate limit non-logged in users because that’s what a lot of automated scraping comes from.