Many projects now launch with promises of verifiable data. Some even store that data permanently. But what is the real value of that verifiable data? Do users verify that data? What if we find the stored data is incorrect? Let's deep dive.
TL;DR
Verifiable ≠ Verified – Data that can be verified does NOT mean it is verified unless YOU verify it.
Long-term storage of verifiable data is wasteful. It wastes resources to keep unverifiable or false data. Users must constantly re-verify it before using it, which is inefficient. Even with a verified data source, users must have a method to verify the verification and actually use it (e.g., KYVE Trustless API).
Users rarely verify – verification is time and resource-consuming, so most users don’t verify even when it’s possible. This is why trustless verified data holds significant added value.
Pre-Validation Is Key – Only trustless, verified data should be stored as a public good. This ensures reliability, reduces waste, and allows for simple, trustless re-verification of the data.
Verifiable means EFFORT, Verified means TRUST(LESS)
It's essential to know the difference between verifiable and verified in decision-making. A poor understanding of these terms can lead to bad decisions. In our industry, it could mean losing money. To illustrate, imagine a marketing campaign for a new smartphone. It claims the phone has a 48-hour battery life. This statement is verifiable. It can be tested by running the phone under normal usage conditions and measuring the battery life.
However, until you test and confirm the claim, it is only verifiable, not verified. If some tech experts or reviewers have confirmed the info, you must still check other sources. You must ensure the information is truly verified. A corrupt group may have posted a fake review for their own interests. Many consumers rely on this distinction before buying. They want to ensure that marketing claims are not just checkable but have been verified by trusted sources.
It's the same with data; verifiable data is a claim that you can verify the data, not that the data have been verified, meaning that you have to do this part of the job on your own, which can be quite tedious and complicated, and if you do it incorrectly, can lead you to believe that some data were correct when they were not or the other way around.
Why it doesn't make any sense to store verifiable data permanently or as a public good
Storing verifiable data forever is a costly inefficiency. The cost of storage persists if the data is later proven false or irrelevant. This means that incorrect or outdated information still occupies space. It requires ongoing support but delivers no long-term value. Also, verifiable data needs constant validation. This forces each new user to check its accuracy. It leads to wasted time and effort.
Over time, verification sources may disappear, making it impossible to confirm or refute older data. The industry should focus on trustless verified data, not on verifiable data. Storing trustless verified data preserves only accurate, useful information, maximizing storage value. Since verification is complete, data users can "trust" the data.
Yet, they should re-validate it by checking the validation layer and the storage layer and fetching the Storage ID and transaction hash from both layers. This process helps reduce redundancy and inefficiency. Trustless verified data is relevant for a long time since a decentralized validation layer confirms it before storage. This approach saves money. It also keeps stored reliable data for future use.
Do you really verify when it's verifiable?
Just because data is verifiable doesn’t mean it’s actually verified—and that’s where the risk lies. Many people use DeFiLlama to check a blockchain's bridge TVL. They never verify the numbers themselves. They assume the data is correct, but if the source is wrong or manipulated, decisions based on it can lead to severe financial losses.
The same happens in daily life—how many of you weigh a product to check if it matches what’s written on the package? Technically, you could, but you don’t because there’s an assumption of trust. For small things, this isn’t a problem. In blockchain and finance, billions move based on data-driven decisions. So, trusting without verifying is a flaw. The only real safeguard is pre-validation. It ensures data is verified before it is stored or used. It must also allow a simple, trustless way to verify that data.
