Who Started the Dark Web? A Brief History

What the Dark Web Actually Is (and What It Isn't)

The dark web is a collection of overlay networks that require specific software, configuration, or authorization to access. You cannot reach it with a standard browser and a search engine. It is not a single website, not a single creator's project, and not synonymous with the deep web.

Many people confuse the deep web with the dark web. The deep web includes all web content not indexed by search engines: your email inbox, online banking portals, subscription databases, password-protected company intranets. The dark web is a small subset of the deep web, deliberately hidden and accessible only through specialized tools such as Tor, Freenet, or I2P.

The internet has three layers:

The dark web was designed so that no one monitoring the network can see who is talking to whom[1]. Traffic is routed through multiple servers and encrypted at each step, concealing both the sender's and receiver's identity. This architecture means the dark web is not one place you visit—it is a method of communication that many independent networks use.

When you hear 'dark web,' you are hearing about the result of decades of work by cryptographers, privacy advocates, and software developers. No single person started it, and no single technology defines it. The tools and networks we call the dark web today emerged from separate projects with overlapping goals: protecting privacy, resisting censorship, and enabling anonymous communication.

The U.S. Naval Research Laboratory and the Birth of Onion Routing

The technical foundation of the dark web began in 1995 at the U.S. Naval Research Laboratory in Washington, D.C. Three researchers—David Goldschlag, Michael G. Reed, and Paul Syverson—developed a new method called onion routing to protect U.S. intelligence communications online[3][1]. Their goal was straightforward: create internet connections that do not reveal who is talking to whom, even to someone monitoring the network[1].

The original purpose was military and diplomatic. Intelligence agencies needed to communicate securely without exposing their operatives' locations or the fact that communication was taking place. Existing encryption protected message content but not metadata—who contacted whom, when, and from where. Onion routing was designed to solve that problem.

How Onion Routing Works

Onion routing sends your traffic through multiple relays, encrypting it in layers at each step. When you connect, your data is wrapped in several layers of encryption—like an onion. Each relay peels off one layer, sees only the next relay's address, and forwards the packet. No single relay knows both the sender and the final destination.

Here is the basic sequence:

1. Your computer encrypts the message multiple times, once for each relay in the path.

2. The first relay decrypts the outermost layer, sees the address of the second relay, and forwards the packet.

3. The second relay decrypts the next layer, sees the third relay, and forwards again.

4. The final relay (the exit node) decrypts the last layer and sends the message to the destination.

Because each relay sees only the previous and next hop, no single point in the network can trace the full path from sender to receiver[1]. Someone monitoring your internet connection sees encrypted traffic going to the first relay but cannot see where it ultimately goes.

The researchers presented their work publicly on 31 May 1996 at the First Information Hiding Workshop in a paper titled 'Hiding Routing Information'[3]. A follow-up paper, 'Anonymous Connections and Onion Routing,' appeared in 1997. These publications laid the conceptual and technical groundwork for every anonymity network that followed.

Onion routing was not released to the public immediately. The Naval Research Laboratory continued to develop the system internally, but the core idea was now documented and available to other researchers. That openness would prove essential when the technology moved beyond military use.

From Military Research to Public Release: The Tor Project

The Naval Research Laboratory's onion routing remained a classified research project through the late 1990s. In the early 2000s, Paul Syverson began collaborating with Roger Dingledine, a recent MIT graduate, to bring the technology to a wider audience[1]. Dingledine named the project Tor, short for 'The Onion Routing,' and was soon joined by Nick Mathewson, another MIT researcher[1].

The team released the first version of the Tor network in October 2002[1]. Unlike the earlier classified prototypes, this version was published under a free and open-source licence, allowing anyone to inspect the code, run a relay, or use the network. By the end of 2003, Tor had about a dozen volunteer nodes—most in the United States, plus one in Germany[1].

This public release marked a fundamental shift. Onion routing was no longer a tool exclusively for intelligence agencies. Journalists, activists, researchers, and ordinary users could now route their traffic through the same anonymity network that had been designed to protect military communications.

The Founding of the Tor Project

In 2006, Dingledine, Mathewson, and Syverson formalised their work by establishing the Tor Project as a 501(c)(3) nonprofit organisation. This structure separated Tor's development and governance from direct government control, even though initial funding came from the U.S. Naval Research Laboratory and later from DARPA (the Defense Advanced Research Projects Agency).

The Tor Project's mission was—and remains—to advance human rights and freedoms by creating and deploying free and open-source anonymity and privacy technologies. The organisation now receives funding from multiple sources, including government grants, private foundations, and individual donations. You can verify current funding sources in the Tor Project's annual financial reports, published on their website.

The transition from classified military research to a nonprofit-run public network took nearly a decade. What began as a way to protect intelligence operatives became a tool used by millions worldwide to resist censorship, protect privacy, and communicate anonymously. The U.S. government funded the initial research, but Tor is now maintained by an independent nonprofit with a global community of volunteers and contributors.

Ian Clarke and Freenet: The First Public Dark Web Network

While the U.S. Naval Research Laboratory developed onion routing in secret, a separate project was unfolding in the open. Ian Clarke, an Irish student at the University of Edinburgh, released the Freenet design paper publicly in July 1999 and launched Freenet version 0.1 in March 2000[2]. This made Freenet the first public peer-to-peer dark web network—predating Tor's public release by more than two years.

Clarke's motivation was clear and political. He stated that his primary goal was to make it very difficult to censor information, because the internet created the potential to censor and monitor people to a degree never before possible[4]. Freenet was designed as a direct response to the threat of government and corporate surveillance, not as a military communications tool.

How Freenet Differs from Tor

Freenet and Tor solve related problems but use fundamentally different approaches. Tor routes your traffic through multiple relays to hide who is talking to whom, but it still connects you to the regular internet. Freenet, by contrast, is a distributed data store. When you publish a file on Freenet, it is encrypted, split into chunks, and stored across many volunteer nodes. No single node holds the complete file, and no node knows what data it is storing.

You retrieve content by requesting a unique key, and the network reassembles the file from fragments scattered across participating computers. Unlike contemporaries such as Napster, Freenet had no central indexes[2]. This design made the network resilient to shutdowns and tracing—if one node disappeared, the data remained available from other nodes.

Here is the practical difference:

• Tor: You browse websites anonymously. Your identity is hidden, but the content you access is hosted on a server somewhere.

• Freenet: You publish and retrieve content stored across the network itself. There is no central server, and no one can take down a file by shutting down a single machine.

Freenet was built for censorship resistance first, anonymity second. Tor was built for anonymity first, with censorship resistance as a secondary benefit. Both networks share the goal of protecting free speech, but their architectures reflect different priorities.

Freenet's Reach and Legacy

The Freenet software surpassed two million downloads, primarily in Europe and elsewhere[5]. Users included journalists in restrictive countries, activists organising protests, and individuals sharing controversial or banned information. Freenet also attracted users who wanted to share copyrighted material or illegal content—a pattern that would later define public perception of the dark web.

Freenet remains active today, though it is less widely used than Tor. The project demonstrated that a public, decentralised anonymity network was technically feasible and that ordinary people, not just intelligence agencies, could build and run such systems. Clarke's work proved that the dark web could exist outside government control, funded and maintained by volunteers who believed in the right to communicate privately.

Other Early Dark Web Technologies and Precursors

Tor and Freenet were not the only projects building anonymity infrastructure in the early 2000s. Several other systems contributed to the dark web's technical foundation, and the broader cypherpunk movement provided the ideological and cryptographic groundwork that made these networks possible.

I2P: The Invisible Internet Project

The Invisible Internet Project (I2P) launched in 2003 as another peer-to-peer anonymity network. Unlike Tor, which was designed to let you browse the regular internet anonymously, I2P focuses on creating anonymous services within its own network. You can host websites, chat servers, and file-sharing applications entirely inside I2P, and all traffic stays within the network by default.

I2P uses a technique called 'garlic routing,' a variation of onion routing. Instead of sending one message through multiple relays, garlic routing bundles multiple messages together and encrypts them as a single package. This makes traffic analysis harder—an observer cannot easily tell how many messages are being sent or which messages belong to which conversation.

Here is the practical difference between Tor and I2P:

• Tor: Optimised for accessing regular websites anonymously. Most traffic exits the Tor network to reach the public internet.

• I2P: Optimised for anonymous services hosted within the network. Traffic rarely leaves I2P, and the network is designed for peer-to-peer applications rather than web browsing.

I2P remains active today, with a smaller user base than Tor but a dedicated community focused on internal services and peer-to-peer communication. It represents a parallel strand of dark web development, emphasising decentralisation and internal anonymity over access to the public internet.

The Cypherpunk Movement and Its Influence

The technical tools that enabled the dark web—onion routing, distributed data stores, garlic routing—emerged from a broader movement that began in the early 1990s. The cypherpunks were a loosely organised group of cryptographers, programmers, and activists who believed that strong encryption was essential to protecting individual privacy in the digital age.

The cypherpunk mailing list, started in 1992, became a forum for discussing cryptographic techniques, privacy tools, and resistance to government surveillance. Key figures included Eric Hughes, who wrote 'A Cypherpunk's Manifesto' in 1993, and Timothy C. May, who outlined a vision of 'crypto-anarchy' where encryption would enable anonymous communication and commerce beyond government control.

The cypherpunks did not build the dark web directly, but their ideas shaped the goals and design principles of every anonymity network that followed. They argued that privacy required technical solutions, not legal protections, because governments and corporations could not be trusted to respect individual rights. This philosophy is visible in the architecture of Tor, Freenet, and I2P—all designed to function without central authority and to resist surveillance even by powerful adversaries.

Public-Key Cryptography and PGP

None of these anonymity networks would have been possible without public-key cryptography, a mathematical breakthrough from the 1970s that allows two people to communicate securely without first sharing a secret key. Public-key cryptography is the foundation of onion routing, garlic routing, and every other system that encrypts data across multiple hops.

In 1991, Phil Zimmermann released Pretty Good Privacy (PGP), the first widely available public-key encryption software. PGP allowed anyone to encrypt email and files, and it became a symbol of the cypherpunk movement's belief that strong encryption should be accessible to everyone, not just governments and corporations.

Zimmermann faced a criminal investigation in the United States because exporting strong encryption was classified as exporting munitions under U.S. law at the time. The investigation was eventually dropped, but the controversy demonstrated the political stakes of making privacy tools public. PGP's release and the legal battle that followed helped establish the principle that individuals have the right to use strong encryption, a principle that underpins the dark web's existence today.

Public-key cryptography enabled onion routing by allowing relays to encrypt messages for the next hop without knowing the full path. It enabled Freenet by allowing nodes to verify data integrity without knowing what the data contained. Without this cryptographic foundation, the dark web would not exist in any recognisable form.

These technologies—I2P's garlic routing, the cypherpunk movement's ideological framework, and public-key cryptography's mathematical tools—did not create the dark web by themselves. They provided the technical and philosophical infrastructure that made anonymity networks possible, practical, and politically defensible. The dark web is not just Tor or Freenet; it is the result of decades of work by researchers, activists, and cryptographers who believed that privacy was worth protecting even when governments disagreed.

The Silk Road and the Dark Web's Inflection Point

The dark web existed for over a decade before most people had heard of it. Tor launched publicly in 2002, Freenet in 2000, and onion routing research began in 1995. For years, these networks were used primarily by activists, journalists, and privacy advocates. That changed in 2011 when Ross Ulbricht launched Silk Road under the pseudonym 'Dread Pirate Roberts'[6].

Silk Road was a hidden service marketplace on the Tor network, primarily for illegal drugs. According to the FBI, Ulbricht created the site in approximately January 2011 and operated it until law enforcement shut it down in October 2013[7]. During that time, the site was used by thousands of drug dealers and vendors to distribute hundreds of kilograms of illegal drugs to well over a hundred thousand buyers and to launder hundreds of millions of dollars[7].

How Silk Road Used Tor and Bitcoin

Ulbricht anonymised Silk Road transactions in two principal ways[7]:

• Tor network: Concealed the IP addresses of the server, vendors, and buyers. No one could trace where the site was hosted or who was accessing it.

• Bitcoin payments: Concealed the identities and locations of users. Transactions appeared as strings of alphanumeric characters, not names or bank accounts.

This combination made Silk Road feel untraceable to many users. You could browse listings, place orders, and complete payments without revealing your identity or location. The site operated openly within the Tor network, with user reviews, vendor ratings, and a customer service forum—features borrowed from legitimate e-commerce platforms.

The Arrest and Conviction

The FBI shut down the Silk Road website and arrested Ross Ulbricht in October 2013[8]. Investigators used a combination of traditional detective work, blockchain analysis, and operational security mistakes by Ulbricht to identify and locate him. The case demonstrated that Tor and Bitcoin provide anonymity, not invincibility—law enforcement can still trace users who make errors or leave identifying information.

Ulbricht was convicted in 2015 and sentenced to life in prison without the possibility of parole. The conviction covered charges including narcotics trafficking, computer hacking, and money laundering. The case became a landmark prosecution, establishing legal precedent for how courts treat darknet marketplaces and their operators.

What Silk Road Changed

Silk Road did not create the dark web, but it made the dark web famous. Before 2011, most news coverage of Tor focused on its use by dissidents and journalists. After Silk Road, the dark web became synonymous with illicit commerce in the public mind. The site proved that anonymous marketplaces could operate at scale, and its success inspired dozens of imitators.

For users already familiar with Tor, Silk Road demonstrated a new application: e-commerce without intermediaries, banks, or government oversight. For the general public, it introduced the dark web as a place where illegal goods were sold openly. Both perceptions are accurate, but neither captures the full picture. The dark web remains a tool for privacy, censorship resistance, and anonymous communication—Silk Road was one use case among many, but the most visible and controversial.

If you want to understand what else exists on the dark web beyond marketplaces, see What Can You Find on the Dark Web? for a broader overview of content and services.

Common Misconceptions About Who Started the Dark Web

The dark web's association with Silk Road and illegal marketplaces has created widespread confusion about its origins, purpose, and structure. Most of what people believe about who started the dark web is incorrect. Here are the five most common errors and the documented corrections.

Ross Ulbricht Created the Dark Web

Many people assume Ross Ulbricht created the dark web because Silk Road was the first widely publicised darknet marketplace. Ulbricht created Silk Road in approximately January 2011 and operated it until October 2013[7], but he did not create the underlying technology. The Tor network, which Silk Road used, was developed by the U.S. Naval Research Laboratory in the mid-1990s[1] and deployed publicly in October 2002[1]. Freenet, another dark web network, launched publicly in March 2000[2]. Ulbricht built a marketplace on existing infrastructure—he did not invent the infrastructure itself.

The Dark Web Is Illegal

The technology that powers the dark web is legal in most countries. Onion routing was created by U.S. government researchers to protect intelligence communications[1], and the Tor Project operates openly as a non-profit organisation. What is illegal is certain activity that occurs on the dark web, such as drug trafficking, weapons sales, or distributing child exploitation material. Using Tor to browse anonymously, access censored information, or communicate privately is lawful. The distinction matters: the tool is neutral, the use determines legality.

The Dark Web Is Huge

The dark web is a tiny fraction of the internet. The 'deep web' refers to all content not indexed by search engines—databases, private accounts, paywalled articles—which makes up the vast majority of online content. The dark web is a small subset of the deep web: networks like Tor and Freenet that require specific software to access. Estimates vary, but the dark web represents less than 0.01% of total internet traffic. Most of the internet you cannot find on Google is not the dark web; it is ordinary private content.

A Single Person or Organisation Owns the Dark Web

No individual or organisation owns the dark web. It is a collection of independent networks—Tor, Freenet, I2P—each maintained by different groups of volunteers and developers. The Tor network, for example, consists of thousands of volunteer-run relays[1], and anyone can operate a relay or hidden service. Freenet distributes data across participating nodes with no central authority[2]. There is no company, government, or person who controls access, content, or infrastructure. The dark web exists because many people choose to run the software, not because someone owns it.

The Dark Web and Deep Web Are the Same Thing

These terms are not interchangeable. The deep web is any content not indexed by search engines: your email inbox, online banking, subscription services, private databases. The dark web is a specific subset of the deep web that requires anonymity software like Tor to access. Every dark web site is part of the deep web, but the vast majority of the deep web is not the dark web. Conflating the two creates the false impression that all private online content is hidden, illegal, or dangerous.

Each of these misconceptions distorts the dark web's history and purpose. The technology was created by U.S. government researchers[1], released as open-source software[1], and adopted by activists, journalists, and privacy advocates long before Silk Road existed[6]. Understanding who actually started the dark web—and why—requires separating the infrastructure from the marketplaces built on top of it.

Timeline: Key Dates in Dark Web History

The dark web's development spans three decades, from classified military research to public anonymity networks and high-profile law enforcement cases. This timeline documents the key milestones with sources for each date, giving you a quick-reference summary of when major technologies launched and events occurred.

Each date represents a documented milestone, not a complete history. Onion routing research began years before the first public presentation[3], and Freenet's design paper circulated in 1999 before the software launched[2]. The gap between Tor's 2002 deployment and Silk Road's 2011 launch shows that the dark web existed for nearly a decade before most people heard of it.

For a detailed explanation of how onion routing works and why it was designed to hide who is talking to whom, see Dark Web Onion: Understanding Its Role.

How to Verify Dark Web History Claims Yourself

You will encounter conflicting claims about who started the dark web, when key technologies launched, and what motivated their creators. Many articles repeat errors or mix up the deep web with the dark web. You can check the facts yourself by consulting primary sources—the original research papers, project documentation, and court records that document what actually happened.

Look for Primary Sources First

Primary sources are documents created by the people who built the technology or investigated the events. For dark web history, the most reliable sources are:

• Original onion routing papers: The first formal presentation, titled 'Hiding Routing Information,' took place on 31 May 1996 at the First Information Hiding Workshop[3]. This paper explains the technical design and motivation before any public network existed.

• Tor Project documentation: The Tor Project maintains an official history page that documents when work began in 1995 at the U.S. Naval Research Laboratory[3], who created the technology[1], and when the network deployed in October 2002[1].

• Freenet's original announcement: Ian Clarke released the Freenet design paper publicly in July 1999 and the software in March 2000[2], making it the first public peer-to-peer dark web network.

These sources tell you what the creators intended, not what journalists or commentators later assumed. If an article claims onion routing was designed for criminals, check the original 1996 paper—it states the goal was to protect intelligence communications[1].

Check Court Documents for Silk Road

Silk Road is the most misunderstood part of dark web history. Many articles claim Ross Ulbricht created the dark web or that Silk Road was the first hidden service. Court documents prove otherwise. The FBI complaint and trial records are public and document that Ulbricht created Silk Road in approximately January 2011[7], nearly nine years after Tor deployed in October 2002[1]. The FBI's press release states that Silk Road was used by thousands of drug dealers to distribute illegal drugs and launder money[7], but it does not claim Ulbricht invented Tor or the dark web.

You can find the FBI's official statement on their website, which confirms Ulbricht operated Silk Road from 2011 until his arrest in October 2013[8]. This timeline shows Silk Road was built on existing infrastructure, not the other way around.

Use Official Project Pages

The Tor Project and Freenet maintain official documentation that explains their history, design, and current status. These pages are more reliable than news articles because they cite specific dates, researchers, and funding sources. The Tor Project's history page confirms that initial work on onion routing began in 1995[3], that David Goldschlag, Michael G. Reed, and Paul Syverson created the technology at the U.S. Naval Research Laboratory[1], and that Roger Dingledine and Nick Mathewson joined the project in the early 2000s[1].

Freenet's documentation, maintained by Ian Clarke, explains that his primary motivation was to make it very difficult to censor information[4]. This context matters: Freenet was designed for free speech, not commerce or illegal activity.

Be Wary of Articles That Conflate Terms

Many articles use 'dark web' and 'deep web' interchangeably, or attribute creation to a single individual. If an article claims 'Ross Ulbricht invented the dark web' or 'the deep web is where criminals hide,' check whether it cites primary sources. Articles that conflate terms or oversimplify history rarely link to the original onion routing papers[3], the Tor Project's timeline[1], or FBI documents[7]. Without these sources, the claims are opinion, not fact.

You should also watch for articles that claim the dark web is 'huge' or 'growing exponentially' without citing traffic data. The dark web is a tiny fraction of internet traffic, and estimates that claim otherwise often confuse the deep web (all non-indexed content) with the dark web (anonymity networks like Tor).

Cross-Reference Dates and Names

When you see a claim with a specific date or researcher's name, cross-reference it against the primary sources listed above. For example, if an article says 'Tor launched in 2004,' check the Tor Project's history page—it confirms the network deployed in October 2002 and became publicly available in 2004[1]. The distinction matters: the technology existed two years before most people could use it.

Similarly, if an article attributes onion routing to a single person, check the original 1996 paper—it lists three creators: David Goldschlag, Michael G. Reed, and Paul Syverson[1]. No single individual started the dark web; it was a collaborative effort across multiple projects and many years.

By consulting primary sources, court documents, and official project pages, you can verify claims yourself and separate documented history from speculation. The facts are public, specific, and consistent across sources—you do not need to trust any single article, including this one.

Common Mistakes and Misconceptions

Assuming Silk Road started the dark web

Many people first heard the term ‘dark web’ during the Silk Road trial, so they assume Ross Ulbricht built the network itself. In reality, Ulbricht launched Silk Road in approximately January 2011[7], while Tor had already been deployed since October 2002[1].

The dark web existed for nearly a decade before Silk Road appeared. Treating the marketplace as the origin of the technology confuses a single application with the infrastructure it ran on.

Overlooking Freenet and Ian Clarke

Most histories focus almost exclusively on Tor, which makes it easy to miss that Freenet came first. Ian Clarke publicly released Freenet version 0.1 in March 2000[2], two years before Tor deployed in October 2002[1].

Freenet was the first public peer-to-peer dark web network, built to resist censorship rather than to enable commerce[4]. If you only read about Tor, you get an incomplete picture of who started the dark web.

Believing one person invented the dark web

Simplified narratives often attribute the technology to a single founder, but the documented history shows otherwise. Onion routing was created by David Goldschlag, Michael G. Reed, and Paul Syverson at the U.S. Naval Research Laboratory in the mid-1990s[1].

Ian Clarke built Freenet separately, and Roger Dingledine and Nick Mathewson later developed Tor with Paul Syverson[1][2]. No single individual started the dark web; it emerged from multiple projects and research teams over several years.

Thinking the dark web was designed for criminals

Media coverage of drug markets and illegal services creates the impression that anonymity networks were built for criminal activity. The original goal of onion routing was to protect U.S. intelligence communications online, not to hide illegal transactions[1].

The 1996 paper ‘Hiding Routing Information’ describes connections that do not reveal who is talking to whom, even to someone monitoring the network[3][1]. The technology predates Silk Road by fifteen years and was funded by the Office of Naval Research[3].

Confusing the dark web with the deep web

The two terms are often used interchangeably, but they describe different things. The deep web includes all non-indexed content—private databases, email inboxes, paywalled pages—while the dark web specifically requires anonymity software like Tor[1].

This confusion leads to inflated claims about the dark web’s size and danger. When you see statistics about the ‘dark web’ being enormous, check whether the source is actually measuring the deep web.

Assuming Tor was the first dark web network

Because Tor is the most widely known anonymity network, people often assume it was the first. Freenet’s public release in March 2000 predates Tor’s deployment in October 2002 by more than two years[1][2].

Freenet also worked differently: it had no central indexes and distributed encrypted data across participating nodes, making it resilient to shutdowns and tracing[2]. Understanding this sequence helps you separate the technology’s origins from its most famous implementation.

Conclusions

• The dark web began as a research project, not a criminal marketplace. Onion routing was created by David Goldschlag, Michael G. Reed, and Paul Syverson at the U.S. Naval Research Laboratory in the mid-1990s[1].

• No single person started the dark web. Ian Clarke released Freenet in March 2000, and Tor deployed in October 2002 as a separate effort[1][2].

• Ross Ulbricht did not invent the dark web. He launched Silk Road in January 2011, nearly a decade after Tor was already running[1][7].

• The dark web and deep web are different things. The deep web includes all non-indexed content, while the dark web requires anonymity software like Tor[1].

• To verify any claim about the dark web’s origins, check primary sources: the Tor Project’s history page, the 1996 onion routing paper, and FBI court documents[1][3][7].

If you want a clear explanation of what the dark web actually is before exploring further, start with Dark Web Definition: What You Should Know.