Proof of Reserves Is Not Proof of Solvency. Here's Why

What does proof of reserves actually prove?

Since the exchange failures of prior cycles, proof of reserves has become a standard trust signal. The typical implementation publishes on-chain wallet balances, sometimes paired with a Merkle-tree commitment that lets users verify their own balance is included in the total liabilities. That is a genuine improvement over blind trust. But it proves a narrow claim: that at a snapshot in time, the exchange controlled certain addresses holding certain assets.

Solvency is a different statement entirely. It requires that verifiable assets exceed total liabilities, including off-chain debts, leverage extended to counterparties, and obligations denominated in assets the proof does not cover. A reserve attestation that shows assets but never rigorously commits to the full liability side is a partial picture, and the market keeps treating it as a complete one.

Where do the standard proofs fall short?

The credibility gaps that persist into 2026:

• Snapshot timing: reserves can be borrowed before an attestation and returned after, so a point-in-time proof can be gamed
• Liability completeness: Merkle proofs cover user balances the exchange chooses to include, not hidden inter-company loans or fiat obligations
• Shared or borrowed keys: two exchanges can, in principle, point to overlapping wallets unless key ownership is independently proven
• Off-chain and derivatives exposure: perpetual and margin liabilities rarely appear in a simple on-chain reserve figure

An attestation answers whether the coins exist. It does not answer whether they are owed to someone else.

How should users read these attestations?

The direction of travel is toward zero-knowledge proof-of-solvency schemes that commit to both assets and liabilities cryptographically, and toward more frequent or continuous attestations that defeat snapshot gaming. Those are meaningful upgrades worth demanding.

Until they are standard, users should read a reserve report as a floor on assets, not a ceiling on risk. The practical signals: how often proofs are published, whether an independent party verifies key control, whether liabilities are committed with the same rigor as assets, and whether the exchange discloses its leverage and lending activity at all. The exchanges converging toward continuous, liability-inclusive proofs are quietly separating themselves from those still publishing a wallet screenshot with extra steps. That divergence, more than any single number, is what the market should be watching.