EN
Sign inGet started
Glossary

Extractive AI

What is Extractive AI?

Extractive AI refers to methods that pull existing information out of texts or documents rather than writing new content. The output comes verbatim or in a fixed structure from the source – such as invoice numbers from PDFs, key sentences for a summary or the relevant passage in answer to a question. Because nothing is invented, the risk of false statements is lower than with purely generative approaches; the two are often combined.

How KI.ordo helps

In KI.ordo flows, the ‘Field from JSON’ building block picks individual values out of structured data, and a template extracts names, companies, places and dates from texts.

Learn more →: Extractive AI

Matching templates

Topics

  • Information extraction
  • Documents
  • NLP

Try KI.ordo

Agents, flows and templates in one workspace – free 24-hour demo.

Start demo
← All terms

Related terms

Named entity recognition (NER)

Named entity recognition (NER) is a natural language processing technique that finds names of people, companies and places, as well as dates or amounts, in text and assigns each to a category. It turns free text into structured data. Example: the sender’s company, contact person, delivery date and invoice amount are extracted automatically from an email and transferred to a CRM.

  • NLP
  • Information extraction
  • +1
Template

Intelligent Document Processing (IDP)

Intelligent document processing (IDP) automatically reads documents such as invoices, contracts or forms, identifies their type and turns the relevant details into structured data. It combines text recognition for scans, natural language processing and validation rules. Organisations save manual data entry and speed up processes such as handling incoming invoices or post. Uncertain fields should be checked by people.

  • Documents
  • Data extraction
  • +1
Template

Text Summarisation

In automatic text summarisation, AI condenses long texts to their key points. Extractive methods copy important sentences verbatim, while abstractive methods rephrase the content; large language models work mainly abstractively. This is useful for minutes, reports or long email threads; for important decisions the summary should be checked against the original.

  • NLP
  • Summarisation
  • +1
Template

Generative AI

Generative AI refers to AI systems that create new content – text, images, audio, video or program code – rather than merely sorting or assessing existing data. They are based on models that learn patterns from large amounts of data and use them to produce plausible new outputs. In business it is used, for example, to draft emails, summaries, product descriptions or image ideas. Results can contain errors and should be checked before use.

  • Core concept
  • Content creation
  • +1