Decoder. plain-English AI glossary

Trust and safety

● Core

Also called T&S

The team and function responsible for keeping a platform from being used to hurt people.

Think of it like

The building management that writes the house rules, handles complaints, and quietly removes the tenants who make life dangerous for everyone else.

Example

A T&S team writes the content policy, builds the tooling to enforce it, reviews escalations, and reports illegal material to authorities.

How it actually works

T&S spans policy (what is allowed), enforcement (detecting and acting on violations), and operations (review queues, appeals, legal reporting). For AI products it also owns model behavior guidelines and abuse response. The job is a constant balance between openness and harm, under real legal and reputational stakes.

For product teams

The function that turns "we should be safe" into shipped policies, tooling, and enforcement.

For engineers

The org that owns content policy, enforcement pipelines, and escalation/reporting workflows around the product.

Related

Read anything AI without the jargon

Look up any term in plain English, or save terms as you read with the free Chrome extension.

Open DecoderAdd to Chrome