The web has a habit that nobody notices until it fails: it asks where a thing lives, not what the thing is. A photograph is whatever sits behind a particular URL. Move the server, let a bill lapse or reorganize a newsroom, and the address can become a tasteful little 404. Protocol Labs began with a different proposition. What if the photograph had an address derived from its contents, and any computer holding the right bytes could answer the request?
That idea became the InterPlanetary File System, or IPFS, the 2015 open protocol most closely associated with Protocol Labs. The name sounds as if it should arrive with freeze-dried ice cream, but the engineering problem is earthly and familiar: links rot, platforms disappear, duplicated files waste bandwidth, and too much of the internet depends on a few owners and data centers. IPFS breaks data into content-addressed blocks, connects them in verifiable structures and lets peers find and exchange them without requiring one canonical server.
Where is it?
server.com/final-v7.pdfWhat is it?
bafy…a fingerprint of the bytesThe distinction is useful, but it does not make storage free or permanent. IPFS is a protocol, not a cloud provider. Somebody still has to keep the bits online. This is where Filecoin enters the picture. The network turns storage into a market: clients make deals with independent providers, while cryptographic proofs help verify that the promised data remains stored. IPFS addresses and moves content; Filecoin supplies an incentive to persist it. They can work together, but neither requires the other.
A toolbox, not a walled garden
Protocol Labs' durable advantage is not one application with a polished login screen. It is a stack of reusable primitives. libp2p handles the graceless chores of peer-to-peer networking - discovery, transports, encrypted connections, stream multiplexing and the acrobatics needed to connect devices behind network barriers. IPLD provides a common data model for links between content-addressed objects. Multiformats gives hashes and identifiers self-describing wrappers so systems can change algorithms without making yesterday's data unreadable.
Archives, wallets, data tools, media, games and distributed applications.
Filecoin coordinates buyers, storage providers, proofs and incentives.
IPFS, libp2p, IPLD and Multiformats address, connect and describe data.
This modularity separates Protocol Labs from conventional cloud companies. Amazon Web Services or Google Cloud sells a managed service with an account, a contract and a vendor relationship. Protocol Labs helped create standards and networks on which many vendors can compete. That can reduce dependence on one operator and make verification portable. It also adds friction. Distributed systems have more moving parts, and IPFS's public routing metadata means privacy-sensitive users need safeguards beyond the protocol's encrypted connections.
Open protocols create a peculiar competitive moat. Any rival can read the code, run a compatible node or fork the project. Protocol Labs cannot rely on secrecy. Its defense is the accumulated work around the primitive: implementations in several languages, testing, documentation, researchers who understand the trade-offs and communities willing to keep the standards compatible. The product is partly software and partly agreement. That agreement is harder to copy than a repository.
“We believe the internet is humanity's most important technology. Our mission is to improve the internet and computing, generally.”Juan Benet, founder and CEO
Who actually uses this?
The customer list resists a neat logo wall because some participants are customers, some are users, some are contributors and some are independent businesses. Developers use IPFS to distribute application data. Blockchain teams use content identifiers for assets and off-chain records. Storage clients and providers meet through Filecoin. Archivists preserve material that should not depend on one server. Researchers move large, verifiable datasets. Enterprises can use peer-to-peer distribution to reduce repeated transfers inside large fleets.
The examples are more revealing than a generic claim about “Web3 adoption.” The Internet Archive has used Filecoin for portions of its digital collection. Starling Lab uses IPFS and Filecoin in systems intended to preserve the integrity of sensitive digital records. Netflix collaborated with the IPFS community while exploring peer-to-peer distribution of software containers. Polygon and Filecoin developed tutorials, grants and hackathons to bring decentralized storage into games, NFTs and other applications. These users are buying resilience, integrity or interoperability - not decentralization as decoration.
The company dissolved into a network
Protocol Labs now describes itself less as a laboratory with several projects and more as an innovation network. Its website counts more than 750 projects and teams, 3,000 active contributors and a combined $40 billion valuation across network companies. The final number is easy to misread. It is not Protocol Labs' valuation. It is a way of describing the economic mass of organizations connected to the network.
The organization wants to cover the awkward distance between a research result and a durable institution. PL R&D funds pre-commercial invention. PL Capital invests when an idea looks capable of becoming a venture-scale company. Accelerators, grants, hackathons and shared platform services help teams recruit, find vendors, meet funders and reach users. In this telling, the company is connective tissue: part research sponsor, part venture studio, part investor and part open-source convener.
That model also explains the business. Protocol Labs is not primarily charging per seat for IPFS. The core code is open source. Financial returns can flow from investments and network-related holdings, while grants and programs push useful infrastructure forward even when its benefits are hard for one company to capture. The exact finances of the private company are not public. The famous 2017 numbers belong mostly to Filecoin's token financing, not to an ordinary venture round: a $52 million advisor sale was followed by a public token sale of about $205.8 million.
A number worth separating
Roughly $257 million refers to the two-part Filecoin token sale. It should not be casually labeled as Protocol Labs equity funding. The distinction matters because token buyers, company shareholders and storage-network participants hold different claims.
Why robots showed up at an internet lab
The network's newer interests can look like a founder's browser with 40 tabs open: digital rights, programmable economies, AI agents, robotics, neurotechnology and brain-computer interfaces. There is a through-line. Protocol Labs specializes in coordination problems where useful resources are scattered among strangers. Filecoin coordinates spare storage. A robotics network might coordinate machines and people to collect training data. Privacy infrastructure might let institutions collaborate without revealing every query. Open funding mechanisms might direct money toward public goods that lack a normal customer.
BitRobot is the cleanest bridge between the old and new Protocol Labs. The project uses a decentralized network and crypto incentives to gather real-world robotics data, a scarce input for embodied AI. PL_Genesis makes the broader thesis explicit. Its 2026 program offered more than $150,000 in prizes across Web3, crypto, AI, robotics and neurotechnology, with promising teams eligible for Founders Forge. The previous season drew more than 1,900 registered builders, produced more than 350 projects and advanced over 20 teams.
The risk is sprawl. A network broad enough to include storage proofs and neural interfaces can lose the crispness of a single mission. Its projects also compete with formidable alternatives. IPFS and Filecoin face centralized clouds as well as Arweave, Storj and Sia. PL Capital competes for founders with crypto funds, accelerators and venture studios. AI and robotics attract better-capitalized laboratories. Protocol Labs' answer is that these categories are less important than the machinery connecting them.
The market position: below the app, above the paper
Protocol Labs fits in the layer below most consumer products and above pure academic research. It makes or funds the protocols, data structures, incentive systems and organizations that other teams can turn into services. That makes comparison difficult but gives it unusual leverage. A successful application may disappear; a useful protocol can quietly become part of many applications, including competitors that never display its name.
For a builder, the practical menu is wider than “use Filecoin.” A team can use libp2p as its network stack, store verifiable objects with IPFS, design links with IPLD, apply to a hackathon, seek a grant or meet investors through the network. For an institution, the value is a set of options around integrity, portability and resilience. None removes the need for product design, security, legal judgment or reliable operators. They change who can provide the underlying service and how independently a user can verify it.
The most consequential Protocol Labs product may therefore be its method. Start with a fundamental weakness, publish a primitive, invite strangers to improve it, add economic coordination where persistence is costly, then let independent teams build the interfaces. It is slower and messier than shipping a proprietary app. It can also produce infrastructure that no longer needs permission from its inventor. For a laboratory that wants to improve the internet, becoming less central is an oddly coherent measure of success.
Explore the network
The official sites below lead to the current network, developer documentation, public code, updates and recorded talks.