Learn how AI moderation works
Short guides on how AI understands text and how moderation works, from the basics to what happens under the hood. Each one ends with a quiz.
11 guides
How AI reads text
- Basics6 min read
What is a text classifier, and how does it decide?
A text classifier reads a sentence and returns a probability for each category. How it is built, what the score means and where it goes wrong.
- Basics7 min read
How language models read text: tokens, embeddings and attention
Computers do not read words. How a language model turns a sentence into tokens and numbers, and how attention lets it understand meaning.
- Intermediate6 min read
Training data and labels: why a model is only as good as its examples
Where moderation training data comes from, why labels are often wrong, what synthetic data and teacher models are, and why licences matter.
Measuring a model
- Basics5 min read
False positives and false negatives in content moderation
Every moderation system makes two kinds of mistakes. What they are, what each one costs a community and how to keep both low.
- Intermediate7 min read
Precision, recall and F1, explained with moderation examples
Accuracy hides the problem. What precision, recall and F1 measure, how to compute them in 30 seconds and which one matters for your community.
- Intermediate6 min read
How to choose a moderation threshold
A score of 0.7 is not a decision. How thresholds work, why the default is only a starting point and a simple method to tune them on your own messages.
AI moderation in practice
- Basics5 min read
Why context matters in chat moderation
The same sentence can be harmless or harmful depending on what came before. How conversation context changes moderation, with real examples.
- Intermediate7 min read
Zero-shot classification: how a model understands rules written in plain words
How can a model enforce “No spoilers” if it was never trained on that rule? Natural language inference, entailment and custom moderation rules, explained.
- Intermediate6 min read
Bias in moderation models: why innocent messages with foreign names get flagged
Moderation models can flag “Mohammed missed the train” as hate. Where this bias comes from, how to measure it and how to remove it.
- Basics5 min read
Human in the loop: how AI and moderators work best together
Automatic moderation is fast, people understand nuance. How to split the work, which actions to automate and which to leave to a person.