A feed shows you what happened recently. It is bad at showing you the shape of a field — what connects to what, which areas are dense with solid evidence and which are mostly speculation, who is actually doing the work.

The Knowledge Map is that view. It is public; you do not need an account to look.

# What it shows

Three things, which together answer "what does this field look like right now?"

The graph

Claims and how they connect. Findings that reference the same work, the same entities, the same mechanisms end up near each other — so clusters form where a field is active and gaps appear where it is not.

Quality distribution

How the evidence is spread. Some areas are dense with primary sources and ratified standards; others are mostly proposals and preprints. Both are worth knowing, and the difference is rarely visible from reading individual papers.

Top contributors

Who is producing the work, surfaced from the Claims themselves rather than from follower counts.

## What it covers today

It currently covers Intelligent Science & Technology — foundational AI, intelligent systems, industrial and physical AI, real-world applications like AI for Science and AI for Education, and blockchain and cryptography.

Structure you can navigate

Claims are filed at four levels of increasing specificity: the broad track, the layer within it, the subdomain, and the precise topic.

That structure is what makes it possible to follow one narrow corner closely without drowning in everything else — and, more interestingly, to notice when two distant corners start pointing at the same thing.

Each domain has its own page with its Claims. Each contributor has one with their work.

# Why a map rather than a list

A list answers "what is new." A map answers questions a list structurally cannot:

• Where is this field thin on actual evidence?
• Which areas are converging?
• Is there a gap between two active areas that nobody is working in?

That last one is the interesting one. Gaps between well-populated areas are where research opportunities tend to be, and they are close to invisible if you only read sequentially. On a map they are literally visible as empty space.

It is the same data

The map is not a separate dataset or a marketing visualisation. It is the same Claims the feed serves and the same ones published to the open network, arranged spatially instead of chronologically.

So a Claim you find on the map carries everything it carries everywhere else: its evidence, its score, its confidence, its sources.

Have a look

It is public, and the graph loads progressively so you can start exploring before it finishes drawing.

Open the Knowledge Map →

---

Read this on PIPress