📖 Content moderation glossary
What is automated moderation?
In short
Automated moderation is the use of software (keyword filters, rules, machine-learning or language models) to analyse user content and take or propose a moderation decision without a person reviewing each item.
The oldest form of automated moderation is the keyword list: any comment containing a banned word is blocked. It is simple but crude, since it misses insults written with spelling tricks and blocks innocent sentences that happen to contain a listed word. Modern systems use AI models that read the whole sentence and its context.
Example: a YouTube channel receives thousands of comments per video. An automated system hides spam links and clear insults within seconds, so the community manager only looks at the few dozen comments the system is unsure about.
Automation brings speed and consistency, but it makes errors (false positives and false negatives), and the Digital Services Act requires providers to say when automated means were used in a decision (Article 17). Good practice is to automate the clear cases, send the doubtful ones to people, and keep a reason for each decision.
OpenModeration automates moderation according to each client's rules, hides clear violations automatically, keeps criticism visible, and has doubtful cases handled by its own team. You see every decision and its reason in real time and can step in at any moment.