I don't think going back to Google because DDG is not large enough to have its own fully independent index is really the right play. Given their reputation and usually lax attitude towards censoring, I'd say this akin to a supply chain attack, not active censoring.
Agreed, I bet as users of ddg we could press the team at duckduckgo to prioritize a fix for this failure mode of censorship.
Maybe they can somehow detect censorship real-time and have their own hot image cache for a certain sized corpus.
This is a fun problem for them to solve for us now that it's a failure mode they didn't fully appreciate until now. They are a search engine aggregator, they need another feed for internet images. This is an engineering problem to do cheaply centrally or distributed.
