Some frauds never retire; they change costume. The gold scam is the oldest thing in my files, and it has come back wearing a token.
The old costume
The classic version has three parts. There is a deposit that is real but marginal: a staked claim in Nevada or Arizona, a concession in Ghana or Peru, a tailings pile somebody mined out decades ago. There is an assay that says otherwise, produced from a salted sample, or by a laboratory nobody can locate, or legitimately, for a different property. And there is a reason the gold cannot be delivered just yet: export permits, a refinery backlog, a customs bond that needs one more payment. Bre-X did it with salted core samples at Busang, and some six billion Canadian dollars of market value went to pennies between March and May 1997. The West African versions do it with photographs of doré bars and something called a government seal. In every version the money is made selling paper against the metal: certificates, forward contracts, promissory notes.
The new one
The 2026 version keeps all three parts and swaps the paper for a token. The claim is tokenized; each coin, the white paper explains, represents a gram of gold in the ground or an ounce in a vault. The assay becomes a PDF in the data room. The reason for non-delivery becomes the blockchain itself: the gold is “locked,” the proof of reserve is a hash, redemption opens “after mainnet.” Yield is offered for staking the token, which converts a mining scam into a Ponzi with better graphics, and because the token trades, the promoter can gesture at a market price as though it were an appraisal of the metal.
Yield is offered for staking the token, which converts a mining scam into a Ponzi with better graphics.
What the record shows
The records, meanwhile, have not moved. Unpatented mining claims on federal land are filed with the Bureau of Land Management and recorded in the county where the claim sits, and the filings show who holds the claim, since when, and whether the annual maintenance fee was paid. A producing mine has state permits, reclamation bonds and a tax history; a vault has a custodian with a name and an insurance certificate you can confirm with the insurer; an assay has a laboratory with an accreditation number. When the promoters cannot produce these, or produce them and they describe a different property, you have learned which of the two things on offer, the rock or the coin, is the actual product.
A mine worth investing in has also been studied, and the studies climb a ladder everyone in the business knows. A preliminary economic assessment, or scoping study, says the deposit might be worth mining, on numbers that by industry rule of thumb could be off by half. A pre-feasibility study narrows the options and is the first point at which anyone may call part of the deposit a reserve. The feasibility study, the bankable one, prices the whole project to within perhaps ten to thirty percent: the mine plan, the mill, the water, the power, the road, the closure bond. Each is signed by a qualified person who stakes a professional registration on it, and for listed companies in Canada and the United States the reports are public, on SEDAR+ and EDGAR. Canada wrote its rule, National Instrument 43-101, in the wake of Bre-X; the SEC followed with S-K 1300.
A white paper that cites “a feasibility study” and cannot produce it has told you something. So has one that produces a scoping study and calls it feasibility, or one that counts inferred resources as gold in the ground. Inferred resources, by rule, cannot be counted as reserves at all; they are the geologist’s educated guess, and the token is priced as though the guess were a vault.
A real mining company, in my experience, talks about ore grade, strip ratio, recovery rate and the price of diesel. A tokenized one talks about tokenomics, community and the roadmap. When a deck has more to say about the coin than the rock, I pull the claim records first.
The bottom line. If someone has offered you a token backed by something in the ground, send me the white paper before you send them the wire. The claim records take an afternoon, and they have never once agreed with the deck. How we run due diligence.
Related: The confession in the commit log · Checking a founder before you invest.
