On Greek Mythology and Trustless Oracles
A story of oracles, blockchains, data, and decentralization.
In Ancient Greece, oracles were revered individuals known for their exceptional wisdom and ability to provide insights and prophetic predictions of the future.
They were believed to serve as portals through which the gods communicated directly with people.
Phythia is the most influential oracle of all time, also known as the Oracle of Delphi (not Delphi Digital).
This woman was essentially the highest authority both civilly and religiously in male-dominated ancient Greece. She responded to the questions of citizens, foreigners, kings, and philosophers on issues of political impact, war, duty, crime, family, laws—even personal issues.
In modern times the role of oracles has evolved, but their primary function remains unchanged: to provide reliable real-world data.
The blockchain oracle problem refers to the inherent limitations of blockchains in accessing external data: blockchains lack the built-in functionality to fetch data or send data to external systems.
The easiest way is to conceptualize a blockchain as an isolated network, with no Internet connection.
To address this limitation, oracles now play a crucial role in connecting Web3 ecosystems to the real world, enabling blockchains to access real-world data sources and perform advanced computations to achieve interoperability with traditional systems.
Oracles are ubiquitous, they currently find applications in decentralized exchanges, money markets, prediction markets, betting, insurance agreements, and more. Imagine you and Bob want to bet some ETH on how hot the temperature will be tomorrow - you’ll need an oracle for that.
This is a huge feat, as oracles essentially facilitate smart contracts and blockchain-based operations by providing essential external data inputs.
Accordingly, oracles are a key infrastructural component.
What are Oracles?
In plain words, an oracle is an infrastructure acting as an intermediary between blockchains and any off-chain data (APIs, cloud providers, payments systems, IoT devices, and so on).
To be able to do so, oracles have an on-chain and an off-chain component.
On-chain: connects to blockchains to monitor requests, broadcast data, sign transactions, and perform computations.
Off-chain: process requests, retrieve external data, send blockchain data to off-chain systems, and perform off-chain computations.
Why do we need Oracles?
Unfortunately, blockchains cannot solve the Oracle problem.
While blockchains provide trustless verification for on-chain data, verifying off-chain data introduces trust assumptions.
Send Bob $1000 once I receive his $1000 bank transfer to my bank account.
What is the price is $TSLA?
What is the temperature today in London?
What is the correct answer? How can we cryptographically prove that something is effectively true?
Different data sources may yield varying results, leading to subjectivity and potential manipulation and irreversible actions on the blockchain caused by smart contracts that lean on data supplied by the (malicious) data sources.
These challenges can undermine the very essence and value proposition of blockchains.
Off-chain data is fundamental to power core smart contract functions: imagine the mess if there were thousands of different data sources, of which most are centralized, that we don’t know if we can trust, or if they are malicious and trying to manipulate smart contracts outcomes in their favor.
This is also the reason why all of those “supply-chain tracking” initiatives are mostly BS - tracking the supply chain is basically worthless if off-chain data is inserted by the entities that monitor the system. The source of data is impossible to verify and if the information is corrupt, then the whole system doesn’t work from the start.
Attempting to address the oracle problem on-chain poses issues of scalability, costs, and security. Embedding oracles directly on-chain could create governance challenges and overload the network with high volumes of data that oracles need to process, which would trickle down all the way and lead to unsustainable transaction costs.
Last but not least, the introduction of oracles on-chain is not feasible from a security point of view, as it would add complexity to the base layer of blockchain networks, compromising their stability and increasing the possible surface for an attack.
To maintain a clear focus on consensus and determinism, oracles operate separately from blockchains.
This ensures that blockchains have a lower attack surface and retain their determinism by maintaining a singular focus on consensus, while oracles have the required flexibility needed to generate determinism from a complex and subjective off-chain world without creating dependencies and limitations that put at risk all other applications.
However, relying on an oracle for off-chain data introduces security concerns if the oracles are not trustless.
Centralized oracles can, in fact, undermine the trustless nature of blockchain networks, as users must trust these oracles to provide accurate and reliable data.
The entire point of a smart contract is to achieve determinism through technological enforcement of the contract’s terms as opposed to probabilistic execution carried out by human enforcement.
And no one wants a single point of failure, or really having to trust anyone.
In addition to the “classic” oracle problem, the development of a multi-chain ecosystem with several L3s on top of L2s on top of L1s widens the role of oracles and poses additional risks.
If anything, the demand for oracles services will only increase.
A model relying on centralized oracles does not fit with the core ethos and value proposition of blockchain networks.
In order to overcome these shortcomings, oracles need to create the same security and reliability guarantees of a blockchain.
What is the way to go forward?
For this part of the article, we use Chainlink as a case study to highlight how they decided to overcome the problem of centralized oracles and the CCIP as a solution for the multi-chain oracle problem.
Linking the World
Chainlink Total Value Secured (TVS) is over $13b, according to DefiLama, excluding CeFi. Alone, it represents half of the TVS by DeFi protocols and secures over 300 DeFi protocols.
If we include CeFi, Chainlink TVS grows closer to $75b.
How did Chainlink manage to become the leading Oracle solution?
Chainlink stands out as the leading oracle solution, seeking to decentralize oracles and achieve determinism on the oracle layer.
Probabilistic vs Deterministic outcomes
Probabilistic outcomes, while theoretically sound, can falter in the face of scams and fraud: you can define agreements and outcomes perfectly, but the system of paper guarantees might fail (scams, frauds, etc.).
This is why for a high-quality and reliable agreement you need both to have:
A definite truth
A system of contracts that guarantees the outcome of the defined truth
Chainlink envisions a world where cryptographic truth underpins all reliable agreements and will become the gold standard of our future financial system.
To achieve determinism on the oracle level, Chainlink has developed Decentralized Oracle Networks (DONs).
Currently, there are more than 1007 DONs.
Instead of relying on a single oracle, Chainlink relies on over a thousand of DONs each specific for a particular use case.
Why can we trust Chainlink data then?
Chainlink’s take on the blockchain oracle problem is to achieve determinism by leveraging several principles to build DONs.
Chainlink DONs are decentralized and open-source. This setup allows anyone to independently verify the reliability and security of the oracle network.
The decentralized nature of DONs also ensures that no single node or data source becomes a single point of failure. Instead, all data is authenticated by multiple nodes, which guarantees timely and tamper-resistant data delivery.
This ensures that the data will be available, delivered on time, and resistant to manipulation.
Users can always confirm the integrity of the network as DONs provide cryptographic proof, signed by the nodes, and stored on-chain.
Moreover, DONs are also blockchain and hardware agnostic. This flexibility allows DONs to grow across multiple chains and to perform more advanced computations leveraging cutting-edge software and hardware, such as zk-proofs.
Chainlink has worked diligently to lower the entry barriers for DONs services, for instance, the introduction of the Off-Chain Reporting Protocol slashed costs by 10x, propelling a rapid growth of DONs right after.
While initially focused on Data feeds, Chainlink’s product suite now includes:
Here are a few use cases that can be unlocked:
Data Feeds: on-chain and off-chain asset pricing
VRF: random number generation for on-chain gaming and prediction markets
Proof of Reserves for CEX, Wrapped Tokens, and Stablecoins
Off-chain data sources for Smart Contract Automation
Easy blockchain integration for enterprises
The image below provides a general (although outdated) overview of the division of clients in the main business areas where Chainlink operates.
Chainlink's core business still remains data feeds, but we can already see how it is gaining traction in other niches.
More Blockchains More Problems
In a multichain world, there are multi-ple problems and they will only get worse.
As the number of blockchain networks increases, so does the demand for robust cross-chain solutions.
One such case is to improve the security and reliability of bridges.
Unfortunately, bridge hacks have been a significant security concern in DeFi.
In 2022, 69% of stolen crypto funds resulted from these hacks, according to Chainalysis.
In a few instances, hackers have targeted bridges by manipulating the oracles from which protocols source their data, like token prices, as well as getting access to the multi-sigs that hold the ability to propagate (malicious) messages across networks (e.g. Ronin, Harmony). This is what Chainlink does with the CCIP, so that this will not be an attack vector anymore.
Furthermore, Proof of Reserves can already be used by bridges to prove their collaterals across networks.
Before Chainlink DONs, centralized price feeds were constantly getting exploited. Currently, the same is happening with bridges - and we are waiting for Chainlink to announce the next industry standard with CCIP.
Chainalysis approximated that in 2022 DeFi protocols lost over $400 million across 41 separate oracle manipulation attacks.
These problems will unfortunately only get more complex: DeFi, dynamics NFTs, on-chain gaming, and on-chain insurance, among others, will keep growing.
Accordingly, the increased multichain and multilevel infrastructure of blockchains requires the development of global standards that set a blueprint for safe and trustless cross-chain interactions.
Though Chainlink did solve the issue of communication between the on-chain and off-chain world, the emergence of the so-called “Internet of contracts” demands secure trustless communication across different blockchain networks, and liquidity to be globally accessible — not just within a siloed single chain.
In the words of Chainlink CEO “connecting them would mean leveling up the whole industry”, with the overall goal of creating a new and more secure financial system.
Linking the Multichain Universe
Enters Chainlink Cross-chain Interoperability Protocol (CCIP).
The importance of the CCIP for Chainlink cannot be overstated.
The team has been working for over 3 years on the CCIP, which is like 5 centuries in the crypto time continuum
Why did this take so long, you may wonder?
Chainlink is not merely trying to connect and bridge all of these networks.
But most importantly they seek to do so while having deterministic guarantees on the immutability and validity of the data.
What is Chainlink CCIP?
At its core, the CCIP is an interoperability protocol with the purpose of:
Facilitate cross-chain real-world assets
Scale cross-chain ecosystems
Bridge private and public blockchains
Enable token transfers across different blockchains
Assist in the creation of cross-chain applications
The CCIP itself is a single and simple interface through which dApps and web3 entrepreneurs can securely meet all their cross-chain needs.
It is not just a fancy word for a bridge to move tokens, but a protocol focused on communication and messaging.
Sure, CCIP will also create many distinct bridges, moving tokens across chains and dApps. However, this is just a limited part of its overall scope, which is to take the industry to the world of cross-chain smart contracts.
With CCIP you will be able to launch one single smart contract composed of multiple on-chain contracts, and multiple off-chain services, providing both bridging in the form of communication and the ability to get data and do trust-minimized off-chain computing.
You’ll be able to leverage some chains for their security guarantees, others for their speed or scalability guarantees, and yet more chains as storefronts.
Imagine how easy would be for developers to deploy tokens or mint NFTs across tens of networks and for users to have a much better wallet UX that abstracts different networks.
The magnitude of this update will benefit crypto immensely: think about the countless development hours saved by not having to build a custom multi-chain solution for developers.
In fact, the introduction of CCIP can substantially reduce development times, making them much closer to Web2: quicker, and cost-efficient.
Aside from greatly improving the development times of Web3 solutions, CCIP also brings security and interoperability benefits.
protocols won’t have to build custom bridges or rely on centralized ones
dApps could seamlessly communicate with each other
The CCIP is like the United States of Blockchains.
They are all brought together under the same rules, can enjoy freedom of movement across borders, and communicate far more easily than before.
With the CCIP Chainlink completes the 3 categories of what its protocol and DONs do:
Provision of Validated Data
Off-chain Computing
Cross-chain communication
Contracts will be able to send each other messages to user their services, tokens across contacts with minimal additional work, giving rise to cross-chain smart contracts that are enabled by highly-validated data and bolstered by the ability to leverage advance computations to provide clear cryptographic guarantees to users.
What’s more about the CCIP is that it will be continually improved and updated to support more blockchains, functionalities, and defense mechanisms.
Aside from being powered by Chainlink DONs, the CCIP leverages additional security via the Active Risk Management (ARM) Network.
Introducing the ARM Network
In plain words, it is a secondary validation service that enhances the security of CCIP by identifying and stopping malicious and anomalous activity.
The Active Risk Management (ARM) network is made of on-chain and off-chain components:
Off-chain: ARM nodes that continually monitor all supported chains against abnormal activities;
On-chain: one ARM contract per supported CCIP chain;
The ARM can either “bless” or “curse” a message by reconstructing a transaction's history (Merkle root) and comparing it with the one DON commits.
If they match, the ARM node "blesses" the root.
However, discrepancies lead to the ARM "cursing" the system, halting operations until any issues are addressed.
As such the ARM network is an additional security guarantee that allows the CCIP to add an extra level of security to cross-chain transactions.
Furthermore, as Chainlink released their staking v0.1 product in December '22 there is speculation that the ARM will also be responsible for LINK slashing in the final staking product.
A Variety of Cross-chain Use Cases
Some of the use cases enabled by the CCIP include:
Cross-chain lending: lend and borrow a wide range of crypto assets across multiple DeFi platforms running on independent chains;
Low-cost transaction computation: offload the computation of transaction data on cost-optimized chains;
Optimizing cross-chain yield: move collateral to new DeFi protocols to maximize yield across chains;
Creating new kinds of dApps: developers can take advantage of network effects on certain chains while harnessing compute and storage capabilities of other chains;
Cross-chain transfers: seamlessly transfer assets and data across blockchains;
Cross-development and collaboration between developers: cross-chain dApps;
For instance, Aave and Synthetix have already adopted the CCIP and have been contributing by beta-testing the new Chainlink features.
Synthetix moved cross-chain for a while as operating their protocol on Ethereum had prohibitive costs.
That’s why they have been using Chainlinks oracles for a long time, favoring a generalized secure and reliable solution that everyone can use, over custom integrations that could not scale.
At the same time, the capabilities of Chainlink DONs post-CCIP will continue to expand to support more use cases.
Those will include more advanced computations:
Automation networks of smart contracts
DECO: allowing privacy-sensitive computations to be performed in oracle networks
Fair Sequencing Service: to decentralize transaction ordering
In turn, these growing capabilities will enable more advanced smart contracts use cases further continuing to improve how Oracle networks provide the key infrastructure to support more trust-minimized applications.
Some Issues and Food for Thought
Connecting the world of blockchains is not an easy feat and comes with its own set of problems.
Historically, concerns have been raised regarding the divergence between the increasing size of Chainlink TVS, going from $7B in early 2021 to ~$75B in 2022 (up >800%), and the market cap of $LINK, which almost did not change at all: from $4.7B in early 2021 to $4B (7.5 FDV) today.
Therefore, many have raised the question of value capture by the LINK token.
While Chainlink is arguably the most important piece of infrastructure in the crypto world, the LINK tokenomics fails to empower token holders from accruing value.
Among the possible solutions that the team is evaluating, there is the introduction of staking for LINK, which could work as a sinkhole and alleviate the selling pressure from network participants.
Another issue contributing to this disproportion is the potential dilution of LINK supply, which is still uncertain.
In fact, from Q2 2023 to Q1 2024 Chainlink is expected to release 7% of the total LINK supply in circulation. This rate of inflation is expected to hold on also for subsequent 12-month periods.
Nonetheless, this % is not set in stone, as this rate is “subject to change based on external factors, such as increased network sustainability from user fees”, raising some concerns about the uncertainty.
Last but not least, Chainlink DONs, numbering over 1000, pale in comparison to over 300k Ethereum nodes.
A more decentralized oracle network could offer increased security and redundancy.
Last but not least, Chainlink is currently a permissioned protocol and Chainlink Labs is a centralized entity.
As such, development is mostly done and directed by Chainlin Labs, raising some issues with regard to centralization.
The team has the vision to eventually decentralize the protocol, which of course will come with execution risks.
Interested in testing the CCIP?
Sign up for Mainnet Early Access
Since July 20, CCIP is available to all developers across five testnets: Arbitrum Goerli, Avalanche Fuji, Ethereum Sepolia, Optimism Goerli, and Polygon Mumbai.
Very thorough. Saved this to come back to this later because I should be working right now!
I'm a sucker for any article that combines ancient history, web3 and graphics.