Glossary
AI email terms, explained
Clear, jargon-free definitions of the AI email terms you'll meet around DraftKite — from auto-drafting and triage to tone of voice, RAG, and GDPR.
AI email assistant
An AI email assistant is software that uses artificial intelligence — typically large language models — to help people manage email. It can sort and label incoming messages, summarise threads, and draft replies in the user's own style, usually leaving the person to review and send. It works inside existing mailboxes like Gmail or Outlook.
Auto-drafting (AI draft replies)
Auto-drafting is the use of AI to automatically compose a suggested reply to an incoming email or message before a person writes anything. The generated draft is saved for review, not sent — a human reads, edits, and approves it, keeping final control over what actually goes out.
Email auto-labeling
Email auto-labeling is the automatic classification of incoming messages, where software reads each email and applies a label, category, or folder — by topic, sender, intent, or priority — without manual sorting. It keeps an inbox organized in real time, so messages that need attention surface instead of piling up unread.
Email automation
Email automation is the use of software to perform email tasks — sorting, labeling, routing, replying, or sending — automatically according to predefined rules or AI, reducing manual handling. It ranges from simple filters and auto-responders to AI assistants that classify incoming mail and draft context-aware replies for a person to review.
Email triage
Email triage is the practice of quickly sorting an incoming inbox by importance and urgency, deciding for each message whether to answer it now, defer it, delegate it, or archive it. Borrowed from medical triage, it prioritizes attention rather than processing mail in arrival order, so the most consequential emails get handled first.
GDPR (for email tools)
GDPR is the EU's General Data Protection Regulation, the law governing how personal data of people in the EU and EEA is collected, stored, and processed. For an email tool it means the contents, sender addresses, and metadata it handles are personal data, so lawful basis, data minimisation, and where the data is hosted all apply.
Human-in-the-loop
Human-in-the-loop (HITL) is an approach to designing automated or AI systems in which a person stays involved in the workflow — reviewing, approving, correcting, or overriding the system's output before it takes effect. Rather than acting fully autonomously, the software proposes and a human decides, keeping judgment and control with people.
Inbox zero
Inbox zero is an email-management approach where you keep your inbox empty — or nearly empty — by processing every message to a decision rather than letting mail pile up. Coined by productivity writer Merlin Mann, the 'zero' refers to the mental attention an inbox demands, not merely the message count.
Knowledge base (for AI email)
A knowledge base for AI email is a curated store of an organization's reference material — pricing, product facts, policies, FAQs — that an AI email assistant retrieves from when drafting replies. It grounds generated answers in accurate, organization-specific information rather than the model's general training alone.
Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG) is an AI technique that pairs a large language model with a search step. Before generating text, the system retrieves relevant passages from an external knowledge source and feeds them to the model as context, so answers stay grounded in specific, up-to-date information instead of the model's memory alone.
Smart reply
Smart reply is an email and messaging feature that uses machine learning to suggest short, ready-made responses to an incoming message. Instead of typing, the user picks one of a few brief suggestions — such as “Thanks!” or “Sounds good” — and sends it with a single tap or click.
Tone of voice (in AI writing)
Tone of voice in AI writing is the consistent personality, style, and register an AI text generator reproduces — word choice, formality, warmth, and rhythm — so that machine-written text reads as if a specific person or brand wrote it. In email tools, it is learned from a writer's own past messages.