Deep web vs dark web comes down to intent. The deep web is everything search engines do not index, such as email inboxes, online banking and company intranets. The dark web is a small, deliberately hidden part of it that needs special software such as Tor. The surface web is what search engines index. The US Congressional Research Service uses the same three layers (updated 2 December 2024).
What is the surface web?
The surface web is the part of the web that has been indexed and can be found through a traditional search engine. A search engine finds pages by following links with automated programs called crawlers. Its own documentation says some pages are blocked by the site owner, others cannot be reached without logging in, and that indexing is not guaranteed even for pages it can reach.
That means “surface” describes how a page is found, not how important or safe it is. A public news site, a company blog and a product page are all surface web.
What is the deep web?
The deep web is web content that has not been indexed, so a normal search cannot return it. According to the Congressional Research Service, it includes private intranets, such as the internal networks of companies, government agencies or universities, and commercial databases. Reaching it often requires you to prove who you are and have permission.
In daily life this is ordinary: your email inbox, your online banking page, a company’s internal wiki and a paid research database are all deep web. Our glossary entry on the deep web gives the short definition.
What is the dark web, and what is the dark internet?
The dark web is the segment of the deep web that has been intentionally hidden. The Congressional Research Service says it refers to sites that users generally cannot reach without special software, and that the publishers of these sites are often concealed. People who ask “what is the dark internet” usually mean the dark web, and official sources use the term dark web, so this post does too. The glossary has a short entry for the dark web.
The same source says it can be reached through anonymizing networks such as Tor (The Onion Router) or I2P (Invisible Internet Project). The Tor Project describes onion services as services that can only be accessed over Tor, and says the special .onion address lets people publish content anonymously. Our entry on the Tor network explains how the routing works.
| Question | Surface web | Deep web | Dark web |
|---|---|---|---|
| Can a search engine index it? | Yes | No | Not by a traditional search engine |
| Is a login or special software needed? | No | Often a login and permission | Special software such as Tor or I2P |
| Typical examples | A public news site or blog | Email, online banking, company intranets, paid databases | Onion services and hidden forums |
| Relation to the others | The indexed part of the web | Everything not indexed | A hidden segment of the deep web |
| Who uses it? | Almost everyone | Almost everyone | Mixed: privacy seekers, journalists and criminals |
Deep web vs dark web: is one part of the other?
Yes. The dark web sits inside the deep web, so everything on the dark web is also deep web, but most of the deep web is not dark web. The table above shows why: a banking page is hidden from search engines for privacy, while a dark web site is hidden on purpose and by design. Both are unindexed, but only one needs anonymizing software to reach.
Is the deep web illegal?
No. Being unindexed is not a crime. Every time you log in to a service, you use the deep web. The Congressional Research Service notes that criminal activity occurs on the surface web as well as on the deep web and the dark web, so the layer a page sits in does not tell you whether it is lawful.
Is the dark web illegal?
The dark web may be used for legitimate purposes as well as for criminal ones, says the Congressional Research Service. It lists lawful uses such as free speech, reaching blocked content, political activism, private discussion of health issues, and journalists working with whistleblowers. It also describes darknet markets where illicit goods are sold, including drugs, falsified documents, hacking tools and stolen personal information, credentials or accounts. In April 2022 the Department of Justice described the seized Hydra market as “the world’s largest and longest-running darknet market”, according to that report. Laws differ by country, and our post on how to access the dark web and whether it is legal covers legality and safety.
How many people use Tor?
No one has a reliable count of dark web users or sites. The same source says data on the size of the deep web and dark web are unclear, but Tor Project metrics exist. Tor is not only used for the dark web, since it also routes traffic to ordinary sites, so these figures do not count dark web users.
- 430,054Mean daily Tor users in the United StatesSource: Congressional Research Service, The Dark Web: An Overview (IF12172), updated 2 December 2024
- 13.7%US share of mean daily Tor users worldwideSource: Congressional Research Service, The Dark Web: An Overview (IF12172), updated 2 December 2024
- 37.7%Germany’s share, the largest of any countrySource: Congressional Research Service, The Dark Web: An Overview (IF12172), updated 2 December 2024
Does anyone watch the dark web?
Yes, law enforcement does. The Congressional Research Service says police can use the dark web for online surveillance, sting operations and anonymous tip lines, and that US federal cases are often multiagency efforts such as the FBI-led JCODE team. It also says anonymity on the dark web is not guaranteed, because researchers and law enforcement keep developing ways to identify hidden services or people. That does not mean someone is checking your company’s data for you.
Why does this matter for a company?
Stolen credentials, customer records and access to networks are sold on dark web markets and forums, and the Congressional Research Service lists the sale of personal information, credentials and accounts under fraud. Your own data does not appear there by itself, so a security team needs a way to find out when it does. Dark web monitoring does that, and our guide Hacker and Leak Forums 101 explains how the forums work. For a dated example of takedowns that did not end the activity, read the BreachForums and DarkForums timeline, and see our lists of dark web marketplaces and dark web search engines for the services that exist.
- Know which layer holds your dataMost of your own data is deep web, such as email and internal systems. Only leaked copies may reach the dark web.
- Watch for your names and domainsTrack brand, domain and staff email mentions rather than site names, which change.
- Treat a claim as a leadCheck whether the data is real and yours before anyone reacts.
- Reset what is exposedChange passwords, end sessions and rotate keys for the accounts in the claim.
Sources
The Congressional Research Service defines the three layers, lists lawful and criminal use, and gives the Tor user figures in The Dark Web: An Overview (updated 2 December 2024). The Tor Project explains .onion sites and onion services and how onion services work. A search engine’s own documentation describes crawling and indexing.
Last checked against sources: 10 October 2026.