Bright Sight - Optimistic Oracle

The first optimistic oracle on Solana

Overview: What is an optimistic oracle?

An Optimistic Oracle is a mechanism to provide data in a decentralized and trust-minimized manner. It operates under the assumption that data is generally provided correctly, and disputes are rare. This mechanism leverages economic incentives to ensure data integrity and accuracy.​

Key Features

1. Optimistic Assumption

The oracle operates on the principle that data providers (known as proposers) will submit accurate data. This is achieved by aligning incentives such that it is economically beneficial for proposers to be honest.

2. Dispute Mechanism

If a data point is incorrect, users can dispute it. When a dispute arises, the oracle enters into a dispute window phase where the accuracy of the data is assessed and users can vote on whether the proposed data is valid or invalid. This mechanism ensures that even though the system assumes honesty, it can still handle dishonesty effectively.

3. Economic Incentives

The optimistic oracle leverages financial incentives to maintain data integrity. Proposers must stake a bond such as USDC are rewarded for accurate data submission. If their data is disputed and voted to be in accurate they can lose their entire bond, while those who disputed and voted against this incorrect data receive rewards. Conversely, penalties are imposed for dishonest behavior.

4. Decentralization

The system is decentralized and permissionless, meaning no single entity controls the data. Multiple participants contribute to and validate the data, enhancing the system's robustness and reliability.​

How It Works

Step 1: Data Submission

A proposer stakes a bond, such as USDC, and submits a data point to the optimistic oracle data request. This data point is then available for use by various smart contracts or DeFi applications.

Step 2: Challenge Period

After data submission, there is a predefined challenge period known as the "dispute window". During this time, anyone can dispute the data if they believe it to be incorrect. If no dispute is raised within this period, the data is considered final and correct.

Step 3: Dispute Handling

If a dispute is raised, a user must stake an equivalent bond and the oracle enters a dispute resolution phase. The resolution process allows anyone with staked tokens to vote on whether the data is correct or not.