Our Investment in Starkware

October 30, 2018 | 4 Minute Read

Today we’d like to announce Multicoin Capital’s investment in StarkWare’s $30M Series A financing alongside a cast of other great investors, including: Paradigm, Sequoia, Intel Capital, DCVC, Wing, Consensys, Atomico, Coinbase Ventures, Collaborative Fund, Scalar Capital, Semantic Ventures, Pantera, and Floodgate.

StarkWare is commercializing a specific kind of zero knowledge proof (ZKP) system known as STARKs. I’ll come back to ZKPs and STARKs in a minute, but first, the team.

Led by Eli Ben-Sasson (Technion), Alessandro Chiesa (UC Berkeley), and Michael Riabzev (Technion) on the technical front, the team is among the best on the planet to pioneer STARKs. Prior to founding StarkWare, the cofounders were founding scientists at Zcash, the privacy-protecting digital currency. They also co-authored the original STARK paper, and have been researching ZKP systems for over a decade. Their first-hand experience in the industry is unparalleled.

Uri Kolodony, StarkWare’s CEO, is also an amazing entrepreneur in his own right. We’ve had the pleasure of working closely with him for about 6 months. He assembled a superb team, attracted an and all-star roster of investors, and set a vision, all while executing his on his product roadmap efficiently. With decades of experiences in a range of startups and consulting roles, he also commands the ability to articulate the power of STARKs such that lay people can understand what they do—a non-trivial feat, having tried several times ourselves. Uri and the StarkWare team have been a pleasure to work with thus far, and we couldn’t be more excited to back them.

So, why do Zero-Knowledge Proofs (ZKPs) matter?

In least words possible: ZKPs are a mind-boggling technology. We talked about them in-depth at our Fall Summit last week. They have the potential to reshape society by redrawing trust lines that are so pervasive and universal that we can’t even see them anymore. The implications of this technology are profound. We will undoubtedly explore the implications of ZKPs in future essays, but first, a primer.

ZKPs allow anyone to prove with absolute certainty that they ran some known function correctly on some piece of data without revealing anything about the data.

This is not intuitive. On first glance, it doesn’t even make sense, and sounds impossible.

How can someone run a function on a piece of data and prove that she ran it correctly without sharing the input with a 3rd party? This is the magic of ZKPs. If none of that makes sense I would refer you to this post and this post, both great visual explainers.

ZKPs are in use today. The most common form of ZKP is public-key cryptography, which powers key pieces of the internet and all cryptocurrencies. However, these ZKPs are very narrow in scope: they only allow someone to prove that they own a private key that’s associated with some public key. They aren’t generalizable to any arbitrarily complex computation, as I described above.

Generalizable ZKPs are in production today, albeit in much more limited scope. The two most high profile examples are in the privacy-focused cryptocurrencies, Zcash and Monero. Zcash uses a different kind of ZKP known as a SNARK to allow Zcash users to send Zcash around the world privately. Monero uses yet another form of ZKP known as Bulletproofs to accomplish the same task.

Explaining the differences between SNARKs, STARKs, and Bulletproofs is beyond the scope of this post. If you’d like to learn more, this presentation from Eli Ben-Sasson provides a good overview.

While the power to send money anywhere in the world permissionlessly, privately, and without fear of censorship is a profound breakthrough for humanity, generalizable ZKPs have applications even beyond that.

As of today, it’s clear that there are at least four major applications of ZKP technology. I’m sure we’ll discover more as this technology matures.

  1. Privacy – allow anyone to send money anywhere in the world, privately. This represents a massive breakthrough for human rights around the world. For the first time in human history, people will be able to private store and spend self-sovereign money that governments cannot trace or manipulate.

  2. Blockchain scalability – ZKPs can be used to help scale blockchains by many orders of magnitude. There are two general approaches here: off-chain batching, and recursively composing ZKPs. Of the five fundamental ways to scale blockchains, the ZKP-based approaches can deliver the best results.

  3. Powering off-chain computation in the Web3 Stack – most of the building blocks on the right hand side of the Web3 Stack (see illustration in link above) can use ZKPs to guarantee correctness.

  4. Financial services – in the wake of the financial crisis of 2008, it’s imperative that the financial institutions that power the global economy adhere to risk thresholds. Using ZKPs, these institutions will be able to prove that they’re adhering to risk thresholds to auditors, governments, customers, and the public at large in real time without having to share confidential information on their books. While this is a ways off, we will see nearer-term versions of this rather quickly. The most likely instantiation of this is that a cryptocurrency exchange will prove solvency using ZKPs.

For those who follow ZKPs, there’s a common conception that STARKs are years away from commercialization. These people point at the evolution of public-key cryptography and SNARKs, which indeed took years to commercialize. But in the case of STARKs and StarkWare, history is a poor guide. The StarkWare team is is working with customers—this week—at Devcon4, and will be announcing more in the coming months.

We’ve spent a lot of time researching all things ZKPs this year. They are an amazing technology, and may be one of the defining technologies of our lifetimes. If successful, they will completely re-shape expectations of trust in all facets of life. Given how many layered and compounding trust assumptions power the world today, it’s impossible to forecast the 2nd, 3rd, and 4th order consequences as this technology permeates the world.

We are incredibly fortunate to have a chance to invest in Uri, Eli, Alessandro, Michael, and the Starkware team to accelerate the adoption of ZKPs into the world.

You can follow their blog here, and learn more STARKs on StarkWare’s website and at zkp.science.