why is my AI content tool suddenly failing to establish proper semantic relevance on new pages?

man, i'm pulling my hair out over here. we've been trying so hard to boost our topical authority, and my AI content tool was finally starting to make some headway. but now, for the past two days, it's completely gone haywire on all new pages we try to generate. this is critical, we had a whole content calendar planned around this working.

it just won't properly identify and integrate key entities anymore. it's like the semantic entity extraction part of its process just... broke. the content it's spitting out is generic, lacks depth, and google is clearly not seeing the semantic relevance we're aiming for. we're even seeing some of our newer pages getting flagged as low quality, which is a nightmare for our domain authority. i've checked the APIs, re-ran training data, everything i can think of, but it's still doing it.

here's a snippet from the internal log i keep seeing:

[ERROR] 2023-10-27 14:35:01 - ContentGenerationService: Failed to establish strong semantic relevance for document ID: CGEN-00123.\n[WARNING] 2023-10-27 14:35:02 - EntityExtractor: Low confidence score for key entity extraction in topic 'SaaS CRM integrations'.\n[INFO] 2023-10-27 14:35:03 - KeywordDensityAnalyzer: Detected suboptimal keyword distribution, potential for keyword stuffing or lack of LSI terms.

has anyone else faced this? specifically, issues with their AI tools suddenly failing on semantic relevance or entity extraction? what steps did you take to debug it? could it be a change in google's algorithms that's making my AI's output look bad, or is it definitely something wrong with my tool's internal logic?

i'm completely stuck and running out of ideas. help a brother out please...