TACITUS is under development. The app is experimental. Availability
TACITUS

Small knowledge graphs for AI work.

Keep the sources, qualifications and decisions behind your work connected. Package them as reusable context for your next AI conversation.

Pixel-art diagram. A highlighted source passage connects to a claim and its qualification. The selected records travel together into a context packet that guides a draft.
  1. Source passages
  2. Selected graph
  3. Context for the draft
Ontology
Defines links such as supports and qualifies.
Semantic layer
Selects the records and links relevant to this question.
Capsule and artefact
Carry the selected context and the structure for the output.

Reusable team context

Share the context behind the work.

Save what your team knows and how it approaches the work. Share the source editions, instructions and templates together so a teammate can use the same context in their own conversation.

Capsules

Talk through how the work is done. Save the background, working rules and sources in an editable capsule.

Artefacts

Attach a template, a good example or earlier work to guide the format and level of detail.

Reusable sets

Save the materials you use together. Share the exact versions with your team or export them as a JSON bundle.

The working TACITUS context preview showing an attached capsule, a captured discussion and exact source editions
Working TACITUS prototype in a test workspace. The demonstration discussion and its capsule are attached to a context preview. Hosted generation is paused.

How it works

Read the full FAQ ↗
What goes into a small knowledge graph?

A small knowledge graph links the records that matter to one question. A claim can point to the passage that supports it, the edition it came from, and the condition that qualifies it. A reviewer can follow those links and record where the evidence is incomplete before the team reuses the graph.

What are ontologies and semantic layers?

An ontology gives the team a shared vocabulary for record types and relationships. A semantic layer applies that vocabulary to a task. A source-checking layer follows claim-to-passage links. An update layer follows version changes and keeps the qualifications relevant to the current question.

Can this reduce unsupported claims?

A capsule can supply exact passages, the limits of their support and instructions to flag gaps. ALCE found that strong models frequently left claims without complete citation support. TACITUS uses claim-level review to test whether the cited passage supports the answer and whether important claims have citations.

Can a smaller context make generation more efficient?

Selecting the needed passages can reduce the number of input tokens. Liu and colleagues found that several long-context models performed worse when the relevant evidence was buried in the middle. Measure preparation, retrieval, generation and reviewer time together to find out whether the workflow saves time or money.

What has TACITUS actually measured?

The reproducible assembly checks measure which source excerpts and qualifications survive context preparation, and how many tokens the result uses. Our earlier 48-response model comparison was mixed. Consistent gains in answer quality, cost and review time still need to be demonstrated.

What can I use today?

Registration is open. AI generation is currently paused. Sign in with Google to use the App Library, save conversations, and attach or export exact versions of context.

Additional project

Windtunnel · Synthetic polling

Compare how modeled audiences respond to policies, messages and product ideas. Inspect the assumptions and decide what to test with people.

Source inspection