You might have heard the term of Zero-Knowledge proof in the past, maybe in a research article or maybe in a proof of concept. But Zero-Knowledge proofs are more than just hype, they are the near future.
This year, in 2025, we have seen AI integrate with many products in a rapid trend. And after seeing the amazing push that it had during the RSA conference, the prediction is that this will be the trend for the foreseeable future. But what does this mean to us (the average human)? Well, with the rapid adoption of AI, the way we do things, as individuals as well as organization-wise, is changing. And as we evolve, so does the threat landscape. Meaning when we invent a system, there will be someone who will figure out how to bypass it, break it, or generally misuse it.
And this is the case with Identities. Just a couple of weeks ago, we could read in an article from Viktoria Soltesz how an AI-generated passport passed some checks on a Financial Institution [1]. This was only a Proof of concept; nevertheless, it still holds true. Financial institutions are supposed to keep our money safe, and we expect them to hold some of the highest standards of security. But no system is 100% secure.
The problem then becomes how can we ensure our identity in a reliable way without having to share all our private documents with every single company? The answer is simple: Zero-Knowledge proofs.
What is a Zero-Knowledge proof?
It’s basically something you can prove without actually providing the information. One of the first definitions for a Zero-Knowledge proof was given in the 1985 paper “The knowledge complexity of interactive proof systems [GMR85]” :
“A zero-knowledge protocol is a method by which one party (the prover) can prove to another party (the verifier) that something is true, without revealing any information apart from the fact that this specific statement is true.” In order to understand this concept, we can use two examples:
The most extended example I have heard is as follows: We have 2 actors, one of them wants to prove to the other that they own the key to a door that separates 2 rooms on the same “cave” as you can see in the image below.
In order to be a Zero-Knowledge verifier, the actor that wants to verify the claim asks the prover to go through Room A and exit through Room B. If this is done successfully, we can conclude that the prover actually owns the key to the door between the rooms. But if there is any question about it, we can ask to repeat a similar test, like going through the reverse path.
Another example is :
We have two actors who claim to belong to the same club. In order to prove that both belong to the club, one of the actors places a message in the club safe. If the other actor belongs to the club, they can unlock the safe and deliver the message to the first one. If they fail to do so, the second actor won’t be trusted as belonging to the club.
Some of the problems that this technology brings forth are the fact that now, for every person who wants to ensure that a statement is true, you have to go through a series of actions to demonstrate. This has been solved by creating something called Non-Interactive Zero proofs.
This kind of proof does not require performing an action in order to prove someone’s claim, and this simplifies the process. There are mainly two kinds of non-interactive zero proof techniques:
SMARKs (Zero-knowledge succinct non-interactive argument of Knowledge):
They use elliptic curve pairing and are usually smaller or quicker to verify than their counterpart. But it has a couple of drawbacks; it does require the establishment of a shared key between the prover and the verifier. Anyone with possession of the key will be able to check all the proofs on the system, including generating new ones. They are not quantum-resistant.
STARKs (Zero-Knowledge Scalable Transparent Argument of Knowledge)
Uses polynomials in order to ensure communication. There is no need to create a shared key as there is no need to have a shared “safe space” to store the proofs, hence calling the system Transparent. It is considered post-quantum safe. Nevertheless, it is not really scalable as the proofs require larger computing, but they behave better with larger amounts of data.
Why is Zero-Knowledge Proof more than just Hype?
To sum up, this is a technology that was theorized in the mid-80s but that has been extended in the current internet thanks to its use in cryptocurrencies. Coins such as Monero and ZCash use Zero-Knowledge proofs to maintain transaction privacy for their users. Nevertheless, they are not the only ones that are jumping on the Zero-Knowledge Wagon.
Google is taking a significant step toward enhancing user privacy by adopting zero-knowledge (ZK) technology for private age verification[2], signaling growing mainstream interest in ZK systems. In the broader ZK landscape, several major developments have unfolded recently. Scroll has become the first fully EVM-compatible ZK rollup to reach stage 1 of Ethereum’s rollup decentralization framework, marking a milestone for scalability and security. Meanwhile, Succinct has made headlines with its cutting-edge work on SP1, a high-performance proving system that’s pushing the boundaries of ZK efficiency. On another front, Aztec has launched its public testnet, opening the door for developers to explore privacy-focused smart contracts on its hybrid ZK rollup.
References:
[2] https://www.cryptopolitan.com/google-wallet-introduces-zk-proofs/
