Our Method · The Fable Engine
A hundred years of prospectors missed it. The machines didn't.
01 / Folklore vs. Data
Our proprietary prospect models read the district's complete geoscience record — every historical assay, drill hole, geophysical survey and exploration report filed since the field's first workings — and ranked the ground on evidence alone. The lodes beneath old Brolga had a century of folklore saying they pinched out at the water table. The data said otherwise. [XX,XXX] metres of drilling later, the data is winning — high-grade lodes where the maps said "worked out".
02 / The Four Steps
Same engine. Every commodity.
Read everything
The field's complete record — surviving stope plans, hundred-year-old assay ledgers, modern geophysics — normalised into one statistically honest dataset.
Screen without prejudice
No district folklore. The model doesn't know the field was "worked out" — it knows what the assays say.
Prove the machine before trusting it
The gold model re-discovered the state's known fields blind before it was allowed to rank Brolga. Control recovery is our standard of proof.
Then send geologists
Every target passed the human verification ladder before the first hole was collared — and the same engine now drives near-mine and regional targeting as results land.
AI finds addresses, not orebodies. Machine outputs are internal decision tools — every result, resource and target we publish is reported in accordance with the JORC Code (2012) and stands behind a named Competent Person, never behind an algorithm.