Re-examining the Oracle Machine: Current Situation and Future Opportunities

With Ethereum founder Vitalik Buterin recently exploring the issue of the ultimate Oracle Machine in an article, Oracles have once again become the focus of attention in the crypto community. Currently, Oracles play an irreplaceable role in various fields such as insurance, finance, random prediction, and the Internet of Things. As long as off-chain data retains value, middleware like Oracles will continue to play a key role in the entire Web3 ecosystem. This is specifically reflected in the following aspects:

• With the continuous development of the Web3 ecosystem, the increasing number of dapps and platforms will drive the demand for Oracle Machine data sources and APIs. In the future, dapps, especially DeFi exchanges, are likely to integrate multiple Oracle Machines, using one as the primary data source and 1-2 as backup for cross-validation ( to obtain off-chain metadata, thereby avoiding significant losses caused by the delay or attack of a single Oracle Machine.

• In terms of investment opportunities, although it is difficult to break Chainlink's monopoly in the traditional DeFi data pricing field in the short term, there is still competitive space in the remaining approximately 10% of the market. In addition, there have been some noteworthy innovative protocols and narratives emerging in the subfields of L2, credit, NFT, DID, etc.

• Based on the experience of DEX and L1, looking from the perspective of incremental market demand and L2/ZK as well as emerging application scenarios, the Oracle Machine track is unlikely to be monopolized by a single winner in the long run. There is still enormous development potential and market space worth exploring in areas such as off-chain computing scenarios, pricing of non-standardized on-chain assets like NFTs, and deep integration with AI/ML.

Oracle Machine Overview

) concept

The Oracle Machine is seen as a bridge connecting on-chain and off-chain data, serving as middleware to provide reliable off-chain data services for blockchain projects. The main reason we need the Oracle Machine is that the trust generated by the blockchain itself is currently insufficient to support all the needs of upper-layer applications, thus requiring the Oracle Machine to inject more trust:

• Currently, there is still a huge gap between the on-chain and off-chain data volumes. Compared to the off-chain real world, which has rich data types and a large amount of data, the data volume in the on-chain world is far from sufficient to support its complete independent development away from off-chain. At present, the vast majority of protocols find it difficult to survive without connecting to the Oracle Machine.

• This is also why many people refer to the Oracle Machine as the "necessary evil" of blockchain. When the richness of on-chain data surpasses that of off-chain data and becomes sufficiently rich, we no longer need to rely on Oracle Machines or similar relay agents. This is also the fundamental reason why many DEXs and DeFi protocols insist on adopting on-chain native designs without Oracle Machines.

Blockchain is a closed system that cannot directly connect to the internet, and smart contracts cannot directly access deterministic information from the internet and the real world, such as stock prices, exchange rates, election results, etc. Additionally, due to the existence of consensus mechanisms, blockchain also requires a trusted third party to verify the data. Therefore, the Oracle Machine serves more like an intermediary that facilitates trust between on-chain and off-chain.

It is important to emphasize that the key problem solved by the Oracle Machine is not how to obtain information from the real world ###. In fact, anyone can upload off-chain data to on-chain (, but rather how to help the blockchain trust information from the real world ). Even if off-chain data is transmitted to on-chain, trust and adoption cannot be achieved (.

The key to building trust lies in communication, rather than merely connection. Therefore, the intrinsic value of an Oracle Machine depends on its ability to establish communication channels between on-chain and off-chain, that is, the function of injecting trust into off-chain real-world data by forwarding, verifying, and filtering information. Thus, the key to evaluating an Oracle Machine lies in determining its "degree of trustworthiness."

![OKX Ventures latest research report: Rethinking the Oracle Machine, seeing and not seeing])https://img-cdn.gateio.im/webp-social/moments-7815dd362cba417eacfebc2fc7005bb5.webp(

) Category

According to the form, the Oracle Machine can be divided into software Oracle Machines and hardware Oracle Machines:

• Software Oracle Machine: Provides API/SDK and other services to assist protocols in accessing and transmitting data from third-party servers, such as product prices, weather indexes, flight numbers, etc.

• Hardware Oracle Machine: Widely used in the Internet of Things, commonly seen in electronic sensors and data collectors.

According to data sources, it can be divided into centralized Oracle Machines and decentralized Oracle Machines:

• Centralized Oracle Machine: Usually integrates a single trusted third party ### such as government departments, official institutions, reputable companies, etc. ( to provide data. The advantage is that it can separate data from the untrusted operating system of local devices, preventing data tampering and loss. The disadvantage is that a single centralized data source brings the risk of a single point of failure.

• Decentralized Oracle Machine: An oracle with a distributed consensus mechanism, also known as a consensus oracle. It obtains data from multiple external sources, making it more reliable and trustless.

Centralized vs Decentralized:

Centralized Oracle Machines are highly efficient and feasible. Decentralized Oracle Machines, due to their discrete multi-node structure and cross-referencing process, are more trustworthy and safer.

When efficiency is not the primary consideration, centralized solutions are not the best choice. Clearly, the information provided by a single node is likely to be biased and subjective; in this case, authority comes from despotism rather than credibility.

Due to the requirements and concerns regarding trust risk management, most DeFi applications directly adopt third-party decentralized Oracle Machines like Chainlink, rather than building simple centralized Oracle Machines or running independent nodes to transfer off-chain data onto the chain. ) It is reiterated that project parties or individuals cannot gain community trust even if they transfer data to the chain themselves. (.

| Indicator | Centralized Oracle Machine | Decentralized Oracle Machine | |-----|--------|---------| | Data Feed Price | Single Point | Multi-node | | Feasibility | High | Low | | Time Efficiency | High | Low | | Credibility | Low | High | | Risk Tolerance | Low | High | | Scalability | Low | High |

![OKX Ventures Latest Research Report: Rethinking Oracle Machine, Seeing and Not Seeing])https://img-cdn.gateio.im/webp-social/moments-eb2cda29809972e75a9c4712ea44794a.webp(

) Application Scenarios

In 2021, the Chainlink 2.0 white paper first introduced the concept of a decentralized Oracle Machine Network ### DON (. A DON is a network maintained by a group of Chainlink nodes, allowing Chainlink to provide external data to the blockchain through trustless off-chain computation. To realize this vision, Chainlink has launched a series of products and services such as VRF, Keepers, and CCIP, further expanding the application scenarios of oracles in Web3. The table below lists application cases of oracles in different scenarios such as DeFi, NFT, GameFi, Social, DAO, and cross-chain.

| Scene | Specific Requirements | Functional Services | |-----|------|------| | DeFi | Feed Price | VWAP | | DeFi | Fair Ordering | anti-MEV | | NFT | Community Airdrop | VFR | | NFT | Feed Price | TWAP | | GameFi | Random Scene Generation in Games | VRF | | Social | User Mapping, User Profiling | DID | | DAO | Community Airdrop, Community Governance | DID | | Cross-chain | Cross-chain information, asset transfer | Cross-chain bridge |

![OKX Ventures Latest Research Report: Rethinking the Oracle Machine, Seen and Unseen])https://img-cdn.gateio.im/webp-social/moments-7a0791ea4e1614b138ab5523d6f005cc.webp(

) Track Map

We can classify the Oracle Machine from three dimensions:

1. Depending on the specific functions, in addition to the common DeFi Oracle Machine, there are also credit Oracle Machines, NFT Oracle Machines, and identity ###DID( Oracle Machines.

2. According to the data source, it can be divided into first-party Oracle Machines, third-party Oracle Machines, and multi-party Oracle Machines.

3. According to the data processing method, it can be divided into game theory-based Oracle Machine, reputation-based Oracle Machine, staking-based Oracle Machine, cryptography-based Oracle Machine, and aggregation-based Oracle Machine.

![OKX Ventures latest research report: Rethinking Oracle Machine, Seeing and Unseen])https://img-cdn.gateio.im/webp-social/moments-00394e71d87665c5f0dd80ee4b5f8e37.webp(

Investment Opportunity Outlook

) opinion

1. In the traditional DeFi space dominated by data feed-based Oracle Machines, Chainlink's monopoly position with 80-90% market share is hard to break in the short term. However, there remains competition for second place in the remaining approximately 10% of the market:

   • One possibility is to innovate on the consensus mechanism. Vitalik proposed two potential solutions for the future:

     • Based on game theory, combined with PoS/PoW
     • DAO Committee/Decentralized Court Form

   • Another way is to have a significant improvement in cost-effectiveness, such as price feeding or random number generation being cheaper than Chainlink. For example, Redstone and Ontropy claim that their price feeding costs can be 80-100 times lower than traditional solutions. Only by significantly reducing the calling costs can DeFi or gaming projects consider replacing existing service providers like Chainlink.

2. There are potential opportunities in niche areas of Oracle Machine, such as L2, credit, NFT, DID, etc:

   • L2 Oracle Machine primarily targets native solutions for Optimistic/ZK ecosystems, with low latency and security as core focus points, while the cost of price feeding is relatively secondary.

   • The expansion progress of Chainlink on non-EVM chains needs to be considered. If the pricing feed on a certain chain has basically been monopolized by Chainlink, the competition difficulty will significantly increase.

   • L2 Oracle Machine should largely target the high-frequency trading needs of derivative exchanges ###. L2 can provide higher TPS support, thereby supporting on-chain decentralized derivative trading (. Similar to Pyth and Empiric, which offer more refined pricing or have a more robust data integration process ), directly integrating and validating on-chain first-hand data without the need for off-chain nodes ( may give Oracle Machine protocols an advantage.

   • If the tokenomics is designed reasonably and incorporates more practicality, the tokenomics design of )Chainlink was once controversial due to a lack of practicality, but later on, similar liquidity staking + yield protocols like LinkPool emerged, which activated liquidity to some extent (. We believe that the token price of the Oracle Machine protocol may generally exhibit a relatively stable performance.

   • The credit Oracle Machine combined with on-chain credit expansion theory is likely to trigger a boom in the next cycle.

   • The improvement of the on-chain credit rating system is a prerequisite for building on-chain insurance and recovery mechanisms, thus playing an important role in the development of DeFi as a whole. In the above-mentioned track map, we have listed several credit Oracle Machine cases: CreDA, Cred Protocol, LedgerScore, Spectral, Credora, etc.

   • In this specific sub-field, we believe that more attention should be paid to the differences in on-chain data validation and integration among similar products, such as the coverage of on-chain data, data cleansing capabilities, and the accuracy of credit ratings, etc.). Additionally, it would be a good advantage if the team has a strong background in financial engineering or actuarial skills.

   • At the same time, we also noticed that the entire project of bringing the off-chain credit system on-chain is more complicated than expected:

     • It is difficult for a single project to directly link with a country's social credit system. A key question is whether off-chain economic activities can endorse on-chain credit capabilities. However, a solution like Spectral Finance, which has strong business development capabilities and collaborates directly with large U.S. credit rating agencies, may be a good option.
     • On the other hand, the barrier to integrating off-chain public credit data onto the chain is not high. ( Traditional internet cloud teams theoretically have the capability to develop and implement relevant functions ).

   • NFT Oracle Machine can be mainly divided into two approaches:

     • One type is TWAP time-weighted average, and there is also the use of moving averages.
     • Currently, VWAP cannot be realized due to insufficient trading volume. Considering the illiquid and non-homogeneous nature of NFTs themselves, it is also difficult for us to foresee the realization of NFT VWAP in the short term.
     • Traditional DeFi such as Chainlink and DIA