LibraryData and agents2020Design paperCorpus record
UMA Data Verification Mechanism: Adding Economic Guarantees to Blockchain Oracles
UMA. UMA.
An oracle that does not answer every question in real time. It prices the cost of bribing the token holders who would resolve a dispute, and it asks contract designers to keep the profit of a lie below that cost. The draft is an economic argument, not a feed you can read as truth.
UMA's data verification mechanism is an optimistic oracle. A proposed value stands unless it is disputed, and a dispute is resolved by a token-holder vote. The design rule is that the profit from a lie must be smaller than the cost of buying that vote. A contract that can pay out more than the vote costs has stepped outside the paper.
- Evidence
- Primary paper
- Re-measured
- No
- Assumptions
- 4
- Records linked
- 3
Code
Source, repository, and what is actually hosted
- UMA DVM draft v0.2, 2020https://github.com/UMAprotocol/whitepaper/blob/master/UMA-DVM-oracle-whitepaper.pdf
- UMAprotocol/whitepaperRepository taken from the source URL. No other study cites this repository. This is not an audit.
- Full text is not mirrored. Public availability is not a licence to copy the file onto this site.
- Add to a reading list
01 Claim ledger
What the paper is allowed to say
Each row is a sentence already in the study. The status is the same on every row: a model claim, not a live measurement.
Claim 01 · Paper model
Silence accepts
Undisputed proposals become the value. The system is cheap when nobody fights. It is also wrong when nobody fights and the proposal was a lie.
Claim 02 · Paper model
The vote is the backstop
Token holders commit and then reveal votes, so a voter cannot simply copy the emerging answer. That is a mechanism detail, not evidence that voters know the truth.
Claim 03 · Paper model
Cost versus profit
The paper's framework compares the cost of corrupting the oracle with the profit a contract would hand a liar. Security is an inequality about those two numbers.
Claim 04 · Paper model
The contract has to cooperate
The oracle does not automatically know how much a given contract can extract. Designers are told to keep payouts inside the inequality. A logo does not enforce it.
02 Three cuts
Observation, model, falsifier
A desk does not stop at the summary. Each claim is cut three ways, using only this study's own assumptions and checks. Nothing here is a new figure.
01 Silence accepts
Observation
What the study says
Undisputed proposals become the value. The system is cheap when nobody fights. It is also wrong when nobody fights and the proposal was a lie.
Model
What has to hold
Disputes are possible in the window. A censored dispute path collapses the mechanism to 'first proposal wins'.
Falsifier
What would retire it
Is an undisputed value being described as verified truth?
02 The vote is the backstop
Observation
What the study says
Token holders commit and then reveal votes, so a voter cannot simply copy the emerging answer. That is a mechanism detail, not evidence that voters know the truth.
Model
What has to hold
The voting token is liquid enough that the cost of buying a swing is meaningful. The paper does not prove that a market exists.
Falsifier
What would retire it
What does it cost to swing the vote, at a stated token price?
03 Cost versus profit
Observation
What the study says
The paper's framework compares the cost of corrupting the oracle with the profit a contract would hand a liar. Security is an inequality about those two numbers.
Model
What has to hold
The contract's maximum payout is the one used in the inequality.
Falsifier
What would retire it
How long is the dispute window, and can it be censored?
04 The contract has to cooperate
Observation
What the study says
The oracle does not automatically know how much a given contract can extract. Designers are told to keep payouts inside the inequality. A logo does not enforce it.
Model
What has to hold
The contract's maximum payout is the one used in the inequality.
Falsifier
What would retire it
Is an undisputed value being described as verified truth?
03 Sequence
One action, as an operating tape
- 01A contract requests a value and someone proposes an answer with a bond.
- 02During the liveness window, anyone can dispute.
- 03If nobody disputes, the proposal is treated as true.
- 04If someone disputes, token holders vote.
- 05The economic claim is that buying enough votes should cost more than the dispute can unlock.
04 Load-bearing
The argument, and where a pitch drops it
An economic guarantee is not a truth guarantee
The cut
The draft is explicit that public chains cannot fetch off-chain facts, only incentivise reporters. A reader who wants an authenticated data feed is looking for a different design, such as a signed committee, and different failure modes.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
This is an April 2020 draft, version 0.2. Later contracts are not automatically inside its argument.
The token price is inside the security
The cut
Cost of corruption moves when the voting token's price moves. A security argument that treats that cost as a constant has left the paper.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
This is an April 2020 draft, version 0.2. Later contracts are not automatically inside its argument.
05 Register
What has to be true
Model · Not re-measured
The voting token is liquid enough that the cost of buying a swing is meaningful. The paper does not prove that a market exists.
Model · Not re-measured
Disputes are possible in the window. A censored dispute path collapses the mechanism to 'first proposal wins'.
Model · Not re-measured
The contract's maximum payout is the one used in the inequality.
Model · Not re-measured
The 2020 draft is not a later deployment. Later code can add governance, different bonds, or different voters.
06 Divergence
What happened after the paper
Optimistic oracles in this family still use propose-and-dispute. The useful test is unchanged: write down the profit of a successful lie and the cost of swinging the vote. Chainlink is the contrasting design, a committee that publishes a value rather than a vote that might be called.
A later client, parameter set, or reward formula is a different object. Cite this paper for the mechanism. Cite a primary release for the network. This desk has not re-run the proof.
07 Pre-mortem
What to check before you use the idea
- 0 of 4 marked on this browser. A mark is a reading note, not a pass, a rating, or a recommendation.
08 Anatomy
The paper, in the order a builder needs
The problem it names
A smart contract that pays off on an outside event needs a value. A committee can sign the wrong value. UMA asks whether the contract can be designed so that lying does not pay, even if the voters can be coordinated.
What the design proposes
- A price request is proposed and, if undisputed, accepted. Dispute is the expensive path.
- Disputes are resolved by a vote of token holders, with a commit-reveal so votes are not free to copy.
- The design rule is profit-from-corruption less than cost-of-corruption. The contract, not the oracle, is supposed to enforce that inequality.
How the mechanism is specified
- Most queries are meant to settle without a vote. The security argument is about the vote that could be called.
- The cost of corruption is framed as the cost of buying enough voting tokens. That cost moves with the token's market, which the paper treats as an input, not a constant.
- A contract that can pay out more than that cost has left the design rule, whatever logo it uses.
What this page does not treat as proven
- This is an April 2020 draft, version 0.2. Later contracts are not automatically inside its argument.
- An optimistic oracle can be wrong for as long as nobody disputes it. Silence is not evidence.
- Nothing here is a guarantee that a number is true. It is a story about bribes.
Why the desk still reads it
For any contract that cites UMA, write down the most a liar could extract and the cost of swinging the vote. If the first number is larger, the paper's own test fails.
09 Lexicon
Terms, opened into the record
- Cost of corruption
- What an attacker must spend to make the oracle adopt a false value, in the paper's framework.
- Profit from corruption
- What that false value would unlock from the contracts that trust it.
10 Repository
Every linked record on this page
Underlined words open a page that already exists: a concept, a protocol profile, a failure record, or another paper. If a word is not underlined, this desk does not have a record for it.
Concepts
Protocols
This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.
Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.