Filecoin’s Juan Benet on Upgrading the Heart of The Internet with Blockchains

Juan is also the creator of Filecoin, a storage network incentivized by crypto. Both of Juan’s ventures are hugely ambitious as they are in direct competition with the basic rails of the internet, with protocols such as HTTP and data giants like AWS. We begin our conversation with Juan explaining what led him to start IPFS and Filecoin.

Juan points out that all of the layers of the internet that run below the application layer, rely on trust. He believes that the promise of web3 is to be able to upgrade the whole stack with trust-minimized public verifiability primitives.

One of the key innovations of IPFS is in data addressing, which increases the efficiency of data fetching. But there’s more to IPFS than just this, and we talk about some of the other features of IPFS.

Juan says that universities would be a good next step to expand the user base of Filecoin since they deal with large amounts of data and run intensive computations. We go on to discuss Filecoin’s plans to encourage adoption of its technology beyond web3 users.

Listen to the interview in this week’s podcast episode here:

image 1

Watch the video here:

Thanking our podcast sponsors:

  • Built by the team behind Avalanche, Core is an all-in-one command center for Web3 supporting Avalanche, Ethereum, and Bitcoin, a rich ecosystem of dApps, Bitcoin & Ethereum bridges, NFTs, Subnets, and more. Visit core.app to download now.
  • Minima is a new mobile-native layer 1 blockchain. Join over 400,000 node runners on Minima’s Incentive Program, to start earning $MINIMA every day until Mainnet launch. Get started at minima.global/get-involved 
  • Railway is the first DeFi wallet to integrate the powerful RAILGUN privacy system built using zero-knowledge cryptography and available on Ethereum, BSC, and Polygon.


I got started on this path by first thinking about computing and a lower layer.

One of the most significant challenges faced by the scientific community is figuring out how to navigate big data sets and deal with enormous amounts of data. How do you ensure that they are telling the truth? How can you check that they have not been modified, and if they have, how can you keep track of the version control? How can we construct more efficient systems to handle addressing and the movement of the data?

Today, the majority of the internet and the majority of the web are in a very precarious position because the majority of them are driven by centralized corporations and a few entities that are under the control of individuals or a few different kinds of powerful groups that are not optimizing for the success of those platforms.

For the rest of the world, that doesn’t make for a very good system. I think we have a great chance to update the whole computing stack with much better building blocks that can be used to build this kind of international utility.

Imagine having access to computing paradigms, smart contracts, finance, personal storage, communication tools, and so on, with the same level of reliability that you have access to water or electricity in some of the most developed parts of the world. These things are built and kept up by utilities, not by tech companies with very fast growth.

I think we need an international computing platform with those kinds of features. The really big promise of web3 is that it will bring these kinds of “primitive” features to the internet so that we can build a much better stack to make our superpower machine more secure.

The Internet Stack

The decisions that were put together in the seventies and eighties put in place a structure for the internet that defined the development of the whole thing. It is fundamentally insecure and relies on trust, and that is a huge thing that we need to upgrade now.

Once you’ve established a method for reliable, secure private communications between programs, you can build data addressing and data movement protocols on top of that.

In order for us to achieve extremely secure blockchain systems, extremely secure worldwide finance, and a successful superpower machine for everybody, we need to sync the successes of the platform into the lower layers.

The promise of web3 is to be able to upgrade the whole stack with these kinds of public verifiability primitives. 

We have to take it layer by layer and figure out how to make it easy for the public to check. How can we make things safer? How do we make the whole thing more trustworthy? In the past, people have tried to do things the other way around, like start at the bottom and work their way up, but they have not been successful.

In the 1980s, 1990s, and early 2000s, there was a big push to improve the internet from the bottom up, but it never seemed to work. So, you have to start with the application and go from there. This is for adoption reasons, and it’s working, but I feel like we should keep going and upgrade the whole system. 

ISO/OSI layer model consists of a physical layer, and then on top of that, there’s the link layer. The next layer is the network. On top of that comes the transfer layer, and on top of the transfer layer comes the application layer.

All these layers below the application layer rely on trust. Who’s running them?

A lot of different organizations build and maintain these protocols, and they also build and maintain the description of the protocol.

The Internet Engineering Task Force (IETF) is one of the most influential groups in history, just in terms of defining our world today. The IETF started with academic research labs and grew from there to include entire corporations. It formalized how to talk about these protocols and define the specs. It’s been like a 50-year journey to get to where we are.

Much of this was due to massive capital investments from the United States and several European countries, followed by many other nations around the world, which helped seed the construction of the entire internet. This took three to four decades.

Deeper into IPFS and Filecoin

IPFS put in a lot of work to change how data is handled, such as how it is addressed, and you’ve made it so that you can still use the system that the rest of the internet uses to retrieve data in a way that is different from how it usually works. But it’s not just about data addressing, it’s also about how I understand decentralized storage. Most data is now held on different private servers, but you’re looking to change that. How does that piece work?

I wanted to very explicitly separate the notion of incentivizing a community to share resources like storage as a separate piece. That’s where Filecoin comes in. You can think of IPFS as mapping to HTTP and Filecoin as mapping to Amazon s3, Google Cloud storage, or something like that.

IPFS defines the addressing, and IPFS gives you the protocols by which you go and find the content and fetch it and display it locally. A different protocol should address how you store and pay for the storage of the data. That separation is pretty important because you may have a lot of different incentive structures depending on how you decide to source something.

“The big contribution of Filecoin is to organize a cryptocurrency and a blockchain to organize people around the planet to provide that storage.”


A lot of the data in Filecoin is coming from web3. The challenge now is to get web2 to store its data in Filecoin. The next steps are programmability so that people can run machine learning and other stuff with their data, and then computation. That is the layer 2 that should be built out.

There are a lot of really interesting web3 applications that we are supporting and helping grow. The first set of users made sense to be all web3 users that were already using IPFS or that would be able to leverage a lot of the pathway use cases.

Universities are a really interesting set of users for Filecoin because they tend to want to do these massive-scale experiments that produce massive amounts of data. Then they want to run large-scale computation, and because they are not as well funded as large corporations, they are extremely cost-conscious. For them, a cost reduction like what Filecoin is able to give them is very meaningful. They’re willing to trade off missing features over that.

The features that Filecoin offers confer a huge advantage: 

  • Data integrity verification
  • Verifiable proof of the data
  • Select who stores your data
  • Hard storage proof

What’s next?

Governments and universities. 

“We’re pretty close to being able to build a massive scale platform for science to happen, where these massive scale datasets will be stored and computed on.”

What role does the token play?

  • The Filecoin token enables participants to store and hire. It is like the currency that people use to trade with each other on the Filecoin network.
  • It also helps secure the blockchain itself with consensus-oriented collaterals.

“We kind of want to get to a point where all of this can happen under the hood for the users.”

Their applications and their communities can live on whatever chain they want. That’s best for their application and their community. Then they can use Filecoin’s services to store data, compute on data, and so on, using whatever token they want, but it’s being swapped underneath the hood and spent on the network.

I wonder if having to deal with a token might be a barrier for some of the web2 users that you want to onboard. 

Yes. So we have seen that significantly. And so that’s one of the big things for us to solve. One way we’re addressing this today is by having a group of users work directly with source providers. But it’s not a great solution

What we’re thinking about now is how to create these on-ramps, these storage-oriented on-ramps, for large-scale users that pay in fiat and then convert to Filecoin underneath the hood. This is how web3.storage already works.

So imagine having a storage company that looks just like any other web2 storage company, and you hire it using traditional structures, but then, underneath the hood, it is converting your USD or  Euro wire by using Filecoin. That’s like an adoption path next year and the year after. 

For the Internet, it took decades to build these infrastructure layers. You’ve only been building for a few years, but you’ve already made a lot of progress. Juan, thank you so much for taking the time. I really enjoyed getting very deep into the protocol layer of the internet, and I just like this view of how web3 can really replace all of these layers. I had the idea that blockchains were this value layer on top of all these protocols, but I never really questioned that maybe these protocols themselves also need upgrading.



, ,