As AI proliferates and things on the internet become easier to manipulate, there is a greater need than ever to ensure that data and brands are verifiable, said Scott Dykstra, CTO and co-founder of Space and Time, on TechCrunch’s Chain Reaction podcast.
“I don’t want to get too religious cryptographically, but we saw it in the FTX collapse,” Dykstra said. “We had an organization that had some trust in the brand, like I had my personal savings in FTX. I trusted them as a brand.
But the now-defunct crypto exchange FTX was manipulating its books internally and misleading investors. Dykstra considers this akin to querying a database for financial records, but manipulating them in one's own database.
And this goes beyond FTX and extends to other sectors as well. “Financial institutions have an incentive to want to manipulate their records…so we see it all the time and it becomes more problematic,” Dykstra said.
But what is the best solution to this? Dykstra believes the answer lies in data verification and zero-knowledge proofs (ZK proofs), which are cryptographic actions used to prove something about a piece of information – without revealing the origin of the data itself.
“It has a lot to do with whether there is an incentive for bad actors to want to manipulate things,” Dykstra said. Whenever there is a higher incentive, when people want to manipulate data, prices, books, finances or more, ZK proofs can be used to verify and recover the data.
At a high level, ZK proofs work by having two parties, the prover and the verifier, who confirm that a statement is true without conveying any information other than its correctness. For example, if I wanted to know if a person's credit score was over 700, if there is one, a ZK proof – a prover – can confirm this to the checker, without disclosing the exact number.
Space and Time aims to be that verifiable computing layer for Web3 by indexing data both off-chain and on-chain, but Dykstra sees its expansion beyond the industry and into others. As it stands, the startup has indexed major blockchains like Ethereum, Bitcoin, Polygon, Sui, Avalanche, Sei, and Aptos and is adding support for more chains to power the future of blockchain. AI and blockchain technology.
Dykstra's most recent concern is that AI data isn't really verifiable. “I fear we will never be able to really effectively verify that an LLM has been completed correctly.”
Some teams are working to solve this problem today by creating ZK proofs for machine learning or large language models (LLMs), but it can take years to try to create this, Dykstra said. This means that the model operator can tamper with the system or LLM to do problematic things.
There needs to be a “decentralized, but globally available, always-on database” that can be created via blockchains, Dykstra said. “Everyone must have access to it, it cannot be a monopoly. »
For example, in a hypothetical scenario, Dykstra said OpenAI itself cannot own a database of a newspaper for which journalists create content. Instead, it should be something that is owned by the community and managed by the community in a way that is easily accessible and uncensorable. “It has to be decentralized, it’s going to have to be on-chain, there’s no way around that,” Dykstra said.
This story was inspired by an episode of TechCrunch's Chain Reaction podcast. Subscribe to Chain Reaction on Apple Podcasts, Spotify, or your favorite pod platform to hear more stories and advice from the entrepreneurs building today's most innovative companies.
Connect with us:
- On X, formerly known as Twitter, here.
- By email: chainreaction@techcrunch.com
techcrunch
