Modalities
Summarization
Compressing a longer text — a meeting transcript, an article, a chat thread — into a shorter version that keeps the key information.
01 ——
In plain English
Summarization is one of the highest-volume LLM use cases. Modern models can summarise meeting transcripts, legal documents, code diffs, chat threads, podcast episodes, and entire books — at varying levels of fidelity.
Two main styles:
- Extractive — pick the most important sentences from the source verbatim
- Abstractive — generate new sentences that paraphrase the source (what most LLM summaries do)
Common formats:
- Executive summary (1–2 paragraphs)
- Bullet points
- TL;DR
- Action items (from meetings)
- Structured fields (decisions, blockers, owners)
Tools built around summarization:
- Meeting notes — Otter, Granola, Fireflies, Read, Fathom
- Document summarisation — Glean, Notion AI, ChatPDF, Humata
- News / feed digests — Particle, Artifact (when it lived)
- Research — Elicit, Consensus, Scholarcy
Where it fails: Long-context summarisation can drop important mid-document content (lost-in-the-middle effect). Critical-stakes summarisation (legal, medical) still needs human review.
02 ——
Related terms
LLM
Large Language Model — the type of AI behind tools like ChatGPT and Claude, trained to understand and generate text.
Natural Language Processing
The branch of AI focused on understanding, generating, and working with human language — covering everything from spell-check to ChatGPT.
Context Window
The maximum amount of text (tokens) an AI model can read and remember at once during a single conversation.
Generative AI
AI systems that create new content — text, images, audio, video, or code — rather than just classifying or predicting from existing data.
AI Assistant
A conversational AI tool that helps users with tasks like writing, scheduling, research, or answering questions in natural language.
Extended Thinking
A model mode where the LLM spends extra compute reasoning through a problem before answering — trading latency for quality on hard tasks.
Back to glossaryLast reviewed May 2026