The Question Isn't the Asset. It's the Ground Around It.

How big can the resource get? What about the belt and the neighbouring geological data? What is the permitting status, and what are the regional trends? Now you can just ask it.

A new way to do origination and screening

The interesting question about an exploration asset is rarely the asset itself.

In this demo, I chose a random asset in WA that's currently being drilled (Comet Vale). The Pulse Intelligence Partner platform asset card carries all the data: tenure, drill results, a 3D model, and all technical reports published to date. Static, these are all pretty useful.

But what I really wanted to know was whether there is resource expansion potential, and that answer does not sit inside the tenement boundary. It sits in the belt around it, in decades of WAMEX reports that were written, filed, and then effectively forgotten or reviewed manually, report by report.

So, I asked Pulse's data the question via MCP (Claude):

It came back with two distinct mineralisation styles, the historical work within 8 km of the project, a view on where expansion might come from, the permitting record across the region, including quotes from operators' own documents, and, importantly, a balanced view which includes the case against the thesis as well.

That last part matters most. A general model will tend to agree with you, because it is a probabilistic token predictor, and you have just told it what you want to hear. This run is closed to everything except Pulse data. And this data can now be compared with tens of millions of geological data points, not just nearby, but globally.

Every word is extracted, validated, structured, dated and traceable back to the page it came from. That is what we mean by creating the Goldilocks conditions for AI.

This example was reactive: a query on a specific asset. The real power is running it proactively: describe your Goldilocks conditions and ask where they exist globally, including tracking tenement expiry and ownership.

Welcome to a new way of working.

Less searching. More strategising.™

---

The asset, and the brief

The asset is Comet Vale, in the Eastern Goldfields of Western Australia, held by Gorilla Gold Mines (ASX: GG8). The answer was never going to sit inside the tenement boundary. It sits in the ground around it, in decades of open file exploration reports, drill logs and technical documents written, filed with the state, and then lost to search.

So I put one brief to Pulse Intelligence through the Pulse MCP: structured asset data, drill results, historical WAMEX documents, technical reports, disclosure and the regional permitting record. The model reads Pulse and nothing else.

What comes back

• Two mineralisation styles, read from the reports rather than assumed.
• The historical exploration work within 8 km of the project.
• A view on where additional resource might come from.
• The regional permitting record, including quotes from operators' own documents.
• The case against the thesis, which is the part most research skips.

Chapters

Time	Section
0:00	Why structured mining data changes what AI can do
0:50	The global map view and the asset card
1:43	The real question is the ground around the asset
2:20	Resource expansion potential and what the belt history says
2:48	The brief: WAMEX, neighbours, global analogues, permitting
3:18	Reading an asset against its neighbours
4:00	Two mineralisation styles and the archive within 8 km
4:55	The case against the thesis, and why general AI agrees with you
5:42	Regional permitting headwinds and operator quotes
6:15	Running it the other way: proactive global target generation

What is WAMEX?

WAMEX is Western Australia's open file mineral exploration archive, the reports licence holders must submit to the state. Roughly 15 million pages of maps, drill logs, assays, cross sections and interpretations. Public, and close to unusable at scale because it is unstructured paper.

Pulse has extracted, validated and structured it so it can be queried and reasoned over. South Australia is live. Queensland is being ingested.

What is the Pulse MCP?

MCP is the Model Context Protocol, the open standard that connects an AI model directly to a data source. The Pulse MCP lets Claude, ChatGPT or your own internal model query Pulse inside your own workflow, restricted to Pulse data, with every answer traced back to the filing and page it came from.

Why general AI gets mining research wrong

A large language model is a probabilistic token predictor. Point one at a folder of raw PDFs and it will tend to tell you what you want to hear, because agreement is the likely next token. That is a confident answer with nothing behind it.

Every word in Pulse is extracted, validated, normalised, dated and traceable to its source filing, and the model works only inside that. Goldilocks conditions for AI.

How is this different from a mining data subscription?

A subscription gives you a screen to read. This is data built to be queried by machines: a platform, a documented API and the Pulse MCP over the same structured record, so the work runs inside your model, your spreadsheet or your own systems rather than a window you copy from.

Who this is for

Corporate development and strategy teams, royalty and streaming portfolios, mining funds and resource asset managers, investment banks and advisory desks, sovereign wealth funds, and exploration teams working historical ground.

Pulse Intelligence is the technology and infrastructure layer for mining data. We digitise, validate and structure NI 43-101, JORC and SK-1300 technical reports, exchange disclosure and historical geological archives, then serve them through a platform, an API and the Pulse MCP. More than 450 investment and audit grade metrics across 29,600+ mining assets: drill results, resources and reserves, economics, ownership, royalties, permitting and people. Every figure opens the original page of the original document in one click. That is AuthentIQ™.