Zero-Knowledge Proofs, Blockchain and the Future of Personal Data in the Age of Agentic AI

Posted by:

|

On:

|

Data is the fuel for AI, and the more personal the data, the better. But people increasingly are disinclined to share their data without some reciprocal consideration, putting a potential lid on AI growth just as the agentic revolution starts. One possible way out of the data doldrums and towards the agentic AI promised land is the use of zero knowledge proofs on a distributed blockchain, which an outfit called Midnight is currently pursuing.

A blockchain is a distributed ledger of records that are linked through cryptographic hashes. Once a record, or a block, is written to the ledger, or the chain, it’s distributed to all nodes in the cluster and can be publicly inspected, meaning it’s effectively there forever in an unalterable form.

That permanence and tamperproof-ness makes blockchain technology viable for a range of use cases. Up to this point, the most popular blockchain use case is creating cryptocurrencies, such as Bitcoin and Ethereum. While the presence of fraud with cryptocurrencies (see: FTX’s collapse) has dampened public opinion, there are many more potential uses that don’t bring the “ick” factor. In fact, some of them sound downright practical and–dare we say it–potentially necessary.

One such use case is the potential for data sharing and monetization through zero-knowledge proofs. A zero-knowledge proof is a mathematically proven way for one party to share one piece of information in a Boolean, yes-no manner, without sharing extraneous underlying detail.

(Dave Hoeek/Shutterstock)

For example, a zero-knowledge proof could be used to grant or deny someone access to drink at a bar. Instead of sharing a government-approved ID card, which has the requestor’s verified age but also a lot of other information that is irrelevant and potentially private, the requestor could submit a zero-knowledge proof that delivers the relevant information for that particular transaction.

This is the approach that an outfit called Midnight is currently developing. Instead of preventing bouncers from the need to obtain GDPR-level consent, however, Midnight is developing a data protection blockchain that utilizes zero-knowledge proofs to enable online businesses and their customers to conduct data-enabled commerce in a precise and trusted manner.

Midnight Head of Product Mauricio Magaldi recently briefed BigDATAwire on the unique capabilities of the company’s technology, and the ramifications it could have for data sharing and data monetization in the age of AI.

“The way we built this is it allows the user or the builder of an application to selectively disclose what pieces of data will be shown to a counterparty,” Magaldi said. “We prevent the data from being written on the blockchain unless it’s really necessary, and what gets written on the blockchain are zero-knowledge proofs.”

The real data resides with the user, on their mobile device on their computer. If the user wants to transact with a company in a mutually beneficial way–say, buying a Mocha Frappuccino or viewing a particular piece of content on the Internet–the business could offer the customer a reward or a discount in exchange for getting access to information about the user. Through the zero-knowledge proof, the user can share just enough data to get the reward, and not a byte more.

The distinction between the data and the information about the data is very important, Magaldi said, “both for self-sovereignty in the sense that now you don’t have to offer your data willy nilly as we did for 20 to 30 years.”

How Midnight would use zero-knowlege proofs to share data (Source: Midnight Nightpaper)

“I think that’s a very elegant approach to data privacy, because it enables businesses to continue to operate without necessarily being in touch with the data, which also saves them a lot of money to in terms of data security and data governance, because they don’t have to hold on to your data,” he continued. “It’s just a proof about it.”

Midnight is owned by Input|Output (IOHK), owner of Cardano, which is one the longest stably running blockchains in the world. IOHK was founded by Charles Hoskinson, who also founded Ethereum, was the founding chairman of the Bitcoin Foundation’s education committee, and established the Cryptocurrency Research Group in 2013.

Midnight is currently developing a testnet, essentially an early version of the blockchain its building to use zero-knowledge proofs to enable commerce and data privacy. The company eventually plans to build a full-blown offering.

Zero-knowledge proofs could be used in all sorts of ways. Games could use it to offer rewards. One early Midnight user built a Battleship type game using zero-knowledge proofs. Banks could also use it Know Your Customer (KYC) settings–not to monetize the wealth of private data they have access to, but to minimize the amount of data they get about customers in the first place.

The tech could also play a role in the emerging agentic AI landscape. Magaldi, who previously was a CDO and has a background in data analytics, sees a way for blockchain and zero-knowledge proofs to grease the wheels for more data sharing in the coming world of AI agents.

“Self-custody data is a big differentiator when we’re talking about public chains,” he said. “You own your data, and the way that AI agents or AI companies will start interacting with that data, which is usually higher quality, is to be able to pay rewards by sharing the revenue with whoever provided that data or access to that data.”

In fact, this is already being done, Magaldi said. People may not be aware of it, but there are some cutting-edge firms that are already using blockchain as the foundation for the data record, and then allowing that record to be consumed by AI agents in a private, secure manner.

“That’s the beauty of zero-knowledge proofs, is enabling data privacy on a public blockchain where others won’t,” Magaldi said “Our founder, Charles Hoskinson, says blockchains can’t keep a secret, but Midnight can, and that’s what we’re building.”

Related Items:

Forrester: Generative AI Dominates Top 10 Emerging Technologies In 2023 And Beyond

Blockchain Starting To Feel Its Way into the Artificial Intelligence Ecosystem

Blockchain Solutions Usher in Era of Trusted Big Data

Blockchain At Heart of Personal Data Monetization Service

The post Zero-Knowledge Proofs, Blockchain and the Future of Personal Data in the Age of Agentic AI appeared first on BigDATAwire.

Posted by

in

Leave a Reply

Your email address will not be published. Required fields are marked *