I mean for example "prompt":"cat","negativePrompt":"underwear","maybeNsfw":true - but it's just a cat, and it doesn't matter what you type in the prompt, response maybeNsfw will always be true. I can use in negativePrompt other hard word and maybeNsfw work correctly, but underwear is a trigger word.
Yes, I only hope that the situation will improve in the future. But now some NSFW content (very explicit) is still allowed through without marking.