Dark web mapping reveals that half of the content is legal

A recent global survey commissioned by the Centre for International Governance Innovation (CIGI) showed that seven in ten (71%) global citizens say the “dark net” – “an area of the internet only accessible via special web browsers that allow you to surf the web anonymously” – should be shut down.

The Centre doesn’t say whether the polled individuals were made aware of what the “dark net” is and is used for (illegal things, but also to keep journalists, human rights activists, dissidents and whistleblowers safe while they are trying to carry out their job/mission) before answering the question. If not, it may be assumed that most of them have heard only about the illegal stuff effected with it.

At this point, though, I have to stop and clarify several things. The dark web exists on a variety of dark nets. Some are small, “private” peer-to-peer networks, others are large networks operated by public organizations and individuals (think Tor, Freenet, I2P). But the terms “dark web” and “dark net” are often (erroneously) used interchangeably.

I’m going to take a wild stab and say that when those who performed the aforementioned survey really meant “dark web” instead of “dark net,” and possibly some of the polled individuals thought the same, but in either case the results would likely be similar, if not the same.

What, exactly, is the percentage of legal and illegal content on the dark web?

Cyber threat intelligence firm Intelliagg and dark net indexing company Darksum have recently released the results of their efforts to map the dark web (in this case, only the Tor network), and they are as follows:

• The dark web is much smaller than commonly thought – e.g. the Tor network, which is currently the largest dark network, has approximately 30,000 .onion addresses active at any one time.
• Some dark web content is clearly illegal (carding sites, pornography, fake documentation services, drugs, financial fraud sites, etc.)
• Of the 29,532 .onion identified during the sampling period – two weeks in February 2016 – only 46% percent could actually be accessed. The rest were likely stort-lived C&C servers used to manage malware, chat clients, or file-sharing applications.
• Of those that have been accessed and analyzed with the companies’ “machine-learning” classification method, less than half (48%) can be classified as illegal under UK and US law. A separate manual classification of 1,000 sites found about 68% of the content to be illegal under those same laws.
• At that point in time, these were the percentages when it comes to the contents of the accessed and analyzed sites:

  

“We believe it is important for the public to gain a better understanding of the contents of the dark web in order for there to be a proper debate about its nature, dangers – and potential benefits,” the companies say. “Misunderstanding about the dark net is rife, and has been fuelled by often misleading media coverage. This, in turn, has influenced policy debates based on incorrect assumptions and hyperbole.”

Their plan is to regularly monitor the situation on the dark web and its evolution. As they noted, an important characteristic of the dark web is that it is transitory and ever changeable.