Industry 03 · Media & Knowledge

Find the evidence behind better stories and knowledge.

Archive search, evidence retrieval and editor-reviewed assistance for newsrooms, publishers, archives and knowledge institutions.

Tagore AI with its Sources sidebar open, listing dated official documents next to the answer they support.
Fig. 1 Tagore AI: the dated sources behind an answer. Factly product. Factly product

01 The information problem

What gets in the way.

Factly has published data journalism on official records since 2016 and built Tagore AI for its own researchers. We know what a reporter needs at 6 p.m.: source, past coverage, exact wording. Our tools find them; the editor decides.

  1. Years of past coverage that is hard to search by topic, claim or entity
  2. Claims that need checking against primary sources under deadline
  3. Publishing systems that treat structured data as an afterthought

02 Who we build for

The teams who feel it first.

Organisations

  • Newsrooms and publishers
  • Media research desks
  • Archives and libraries
  • Knowledge institutions

Roles we work with

  • Editors
  • Heads of product and digital
  • Research leads and standards editors
  • Archive managers

03 What we build

Three places to start.

04 Example workflow

An editor-reviewed story brief.

  1. Step 1: A reporter flags a claim circulating about a government scheme.

  2. Step 2: The assistant searches the archive for past coverage and approved primary sources for the facts.

  3. Step 3: It returns a brief: what has been reported, what the sources say, what remains unverified, each line cited.

  4. Step 4: The editor reviews the brief, opens the sources that matter and assigns the story.

  5. Step 5: The published piece links to the primary sources; the brief stays attached to the story record.

Output

A story brief with its evidence attached; the editorial decision stays with the editor.

05 Why Factly

Built on products we run.

Factly’s own researchers were Tagore AI’s first users; SACH won a Meta Llama Impact Innovation Award and a World Summit Award in 2024.[1][2][3][4][5]

A Tagore AI answer to “How has Indian education policy evolved over the decades?”, structured as headed sections with inline citation markers after each claim.
Fig. 2 Every claim carries a citation marker. Factly product. Factly product
Factly product

Tagore AI

Research answers from official documents, with the sources attached.

Client application: Knowledge assistants over your own approved documents and records.

2.5 million+[6]

official documents

650,000+[6]

data tables

900,000+[6]

fact checks

Also relevant

07 Questions

What Media & Knowledge teams ask us.

Will the tool write or publish stories?

No. It retrieves, compares and drafts research briefs with sources. Writing and publishing stay with your editors.

Can it work with our existing CMS?

Usually, through its APIs or export formats. We confirm the integration points during discovery.

Sources (6)

  1. [1] About Factly · factly.in (opens external site) accessed 2026-11-15
  2. [2] SACH browser extension · factlymedia.com (opens external site) accessed 2026-11-15
  3. [3] Llama Impact Innovation Award winners, 2024 · ai.meta.com (opens external site) accessed 2026-11-15
  4. [4] Announcing the WSA winners of 2024 · wsa-global.org (opens external site) accessed 2026-11-15
  5. [5] Dega: home · dega.factlylabs.com (opens external site) accessed 2026-11-15
  6. [6] Tagore AI corpus figures, provided by Factly (November 2026) · tagore.ai (opens external site) accessed 2026-11-16

Discuss a project

Bring us the research your desk repeats.

Tell us about your archive and how a story moves from tip to publication.