Internet censorship is the control or suppression of what people can read, write, and do online. It can involve the blocking of websites, apps, and social media posts. At its most extreme, internet censorship can block access to all online resources located outside the jurisdiction of the censor, and can even involve disrupting access to the internet altogether.
Internet censorship is a contentious topic. While most people agree that some digital content should be off limits, there’s little agreement about exactly what should be restricted or who has the right to decide.
Internet censorship varies greatly from country to country. In Iceland, there are very few restrictions online, whereas in China people struggle to access any information that could be perceived as critical of the government.
Why Trust Us?
We’re fully independent and have been reviewing VPNs since 2016. Our ratings are based on our own testing results and are unaffected by financial incentives. Learn who we are and how we test VPNs.
What Is Internet Censorship?
Broadly speaking, Internet censorship is when someone controls what others can see and do online. It can take a variety of forms and be implemented by a wide range of institutions.
It is also subjective: what some consider to be censorship may be considered by others as ethical behavior.
Schools, for example, may block access to age-inappropriate content. However, they might also block access to news websites and other important sources of information because they do not conform with the school’s religious views.
Governments often justify internet censorship by arguing that it helps prevent the spread of misinformation or harmful content. However, these justifications are often thinly veiled attempts at maintaining social control.
Widespread internet censorship can have a devastating impact on freedom of expression and other fundamental human rights. It can make it difficult to voice opposition, learn about a government’s failings, or organize political opposition.
Sites that are sometimes censored by governments include:
Independent news publications, including those headquartered overseas.
Search engines, which help people find content that the government may wish to suppress.
File sharing sites, including torrenting sites or Google Docs, which enable people to share information freely.
Blogging platforms, such as WordPress and Medium, which make it easy for people to publish their own views and reports of what they’ve witnessed.
Social media sites, which enable people to share their views and connect with other like-minded individuals.
Internet censorship can be implemented by software developers, app store managers, internet service providers (ISP), or by those working on a country’s internet infrastructure.
In this article, we’ll be mainly talking about government-mandated internet censorship.
What Is the Purpose of Internet Censorship?
Internet restrictions are often imposed to prevent the spread of illegal content. However, governments, social media platforms, and other organizations can abuse their power and censor content that’s critical of them.
Internet censorship can be used to silence the opposition during elections. It can cover up human rights abuses during conflicts, and even disrupt attempts to coordinate relief efforts after natural disasters.
Internet censorship is fundamentally about controlling what people can see and say online. That means it’s often about maintaining control, rather than keeping people safe.
Who Is Responsible for Internet Censorship?
Any organization that provides access to information online also has the power to censor it.
An institution such as a school or college may censor internet content because it is considered unsuitable for young people.
It’s common for school networks to impose restrictions on certain websites.
Social media sites and search engines sometimes remove content for violating their own terms and conditions or because of government requests.
Your Internet Service Provider (ISP) might be required to block content following government orders. If the ISP doesn’t comply, it risks losing its license to operate in that country.
In some countries, government agencies use sophisticated technology to monitor and block access to specific websites, social media posts, and apps.
The Global Impact of Online Censorship
Online censorship has both positive and negative effects.
Here are some positive examples of censorship:
Young people can be protected from content that may be harmful to them, for example if a school or college blocks pornography and violent content.
Organizations can improve their productivity by blocking distracting content, such as social media sites.
Employers can avoid the risk of legal action against them by blocking torrenting.
An organization’s network performance can be protected by blocking torrenting, streaming and other data-hungry applications.
Widespread internet censorship can also have negative impacts for millions of people.
Some negative examples include:
Political Repression: Countries like China and Russia use internet censorship to control citizens and suppress political opposition. Their draconian use of censorship undermines the right to freedom of opinion and expression, and the right to peaceful assembly and association. These rights are protected by articles 19 and 20 of the Universal Declaration of Human Rights.
Economic Consequences: There is a huge economic cost associated with internet shutdowns. Our research found that internet blackouts, social media shutdowns, and severe throttling cost $24.7 billion in 2022 alone.
Overblocking: Permissible content may be blocked in error. We found that UK ISPs are blocking 8,000 websites that are not harmful to children. Services offering support related to substance abuse, domestic violence, and mental health counseling are sometimes misclassified and blocked. Thousands of wedding services, photography websites and sites about VPNs are also blocked.
Types of Internet Censorship (How Internet Censorship Works)
There are a number of techniques that may be used to censor the internet. Each type of internet censorship requires specific technology and has its own potential limitations.
IP Blocking
Every web server has an IP address. IP blocking is a technique where censors generate a blocklist of known IP addresses and reset any connections attempting to connect to them. IP blocking can be used to block websites from entire countries or regions.
This is a relatively rudimentary type of censorship and it doesn’t require particularly sophisticated technology. However, it can cause overblocking, as several sites may share an IP address.
To get around IP blocking, a web proxy or a VPN can be used, if these services aren’t also blocked.
Deep Packet Inspection (DPI)
Data travels across the internet in small sections called packets. Deep packet inspection (DPI) is a technique where the contents of these packets are studied as they travel across the internet.
When content is unencrypted, it’s possible to monitor all the traffic in transit. This allows censors to easily monitor and interfere with peoples’ connections.
If content is encrypted, DPI can still see the metadata, which reveals the type of traffic, the IP address, the amount of data transferred, and the date and time of that transfer. That can sometimes be enough to identify banned activity and block the connection.
DPI can be used to block VPN traffic, for example, unless modern VPN obfuscation is used to disguise it.
OpenVPN Traffic Camouflaged as TLS and TCP
Today’s most advanced censors, including China’s Great Firewall and Russia’s Roskomnadzor, use machine learning-based deep packet inspection trained on millions of traffic samples.
These systems can identify VPN usage from statistical patterns alone, without decrypting any data.
Older obfuscation techniques that produce random-looking traffic (such as obfs4 and XOR scramble) are now reliably flagged by these systems, because legitimate internet traffic has recognizable patterns that random noise lacks.
A complementary technique used alongside DPI is active probing. When a firewall suspects that a server is hosting a VPN, it sends test connections to that server to see how it responds.
Servers that respond like VPN servers are then immediately blocked. This is why commercially operated VPN IP addresses are frequently burned within hours of discovery.
Modern obfuscation protocols counter this by returning an innocent-looking web page in response to probe traffic.
EXPERT ADVICE: In China, authorities have the power to compromise site certificates, so a connection that appears to be encrypted may not be, after all.
DNS Tampering
The domain name system (DNS) is used to convert a website address into the IP address of the server where it’s hosted. It is one of the first steps that happens when trying to access a website.
Your DNS resolver finds out the IP address for a website, so your device can connect to its server.
State agencies and ISPs can block access to websites or divert people to an alternative website by interfering with these DNS servers.
Some networks in China use a technique called “DNS mangling”, which returns a false response if someone tries to access a blocked website. This can also be referred to as “DNS lying.”
There is also “DNS cache poisoning”, which interferes with the response sent by a legitimate DNS server.
Although HTTPS encrypts much of your connection requests, DNS is largely unencrypted. This allows censors to view what website you’re trying to access and block the connection.
DNS tampering attacks like these can be circumvented. It’s possible to use an alternative public DNS server that is unaffected by government or ISP censorship, or set up a Smart DNS connection.
For censors, the problem with techniques such as DNS tampering and IP blocking is that you need to know which IP addresses or domains you want to block. There are too many to draw up a comprehensive list, and new sites are launching all the time.
URL filtering works by checking website addresses for banned keywords. If they are found, the connection is reset.
Censorship techniques based on keywords are particularly prone to overblocking. They may not be able to differentiate between pornography and sex education content, or between sites that promote alcohol abuse and sites that help people with recovery.
The “Scunthorpe problem” is so called because stories about the UK town have sometimes been blocked because the town name contains an offensive word.
Using a HTTPS connection, the censors can see the web domains you visit but not the individual pages you view. Censors have the option to block the entire domain or nothing on it, in that case. In China, entire blogging domains such as WordPress are blocked.
Search Engine Censorship
An alternative approach is to prevent search engines from including web sites in their search results in the first place.
Governments can exert pressure on search engines to remove results they don’t want people to see. For some years, Google worked with the Chinese government, providing a Chinese search engine with some results censored at the government’s request. Google withdrew from China in 2010 and its international search engine is now blocked there.
In Europe, Google blocks illegal content to comply with relevant domestic and international laws. It also allows users to request their information be removed from the search engine following a 2014 ruling by the Court of Justice of the European Union (CJEU). This is often referred to as the right to be forgotten.
Internet Throttling
Throttling is a technique where an internet connection is slowed down. If a connection is slowed enough, it can make the user’s device time out and drop the connection entirely. Internet throttling is sometimes called traffic shaping.
Governments in India, Venezuela and Jordan have all throttled internet speeds during periods of political upheaval. At times, they have targeted specific websites, while at others they have throttled all internet traffic in a particular region.
It has been often used in the past by ISPs in the US and Europe to slow down users who are torrenting, so they do not consume too much network bandwidth.
ISPs in the US and Europe often throttle bandwidth-intensive traffic like streaming or torrenting.
Distributed Denial of Service (DDOS) attacks
A distributed denial of service (DDOS) attack uses lots of devices to flood an internet resource with too much traffic, making it impossible for legitimate visitors to access the server. Often, the devices used for an attack have been compromised and their owners don’t know they’re being exploited.
As a web user, there’s not much you can do if the site you want to access is forced offline. You may be able to look for mirror sites or other places that host the same information, or you might have to wait until access to the site is restored.
Port Number Blacklisting
By blocking certain ports, censors can block entire applications. For example, web access, email, instant messaging, or VPN traffic.
This approach is sometimes used in the workplace to stop unauthorized applications, such as torrenting.
On-Device Software
In some cases, software can be installed on a device to censor the content that can be viewed on it. Most commonly, this is used by parents to protect children and young adults from content that is harmful to them.
In 2021, the Lithuanian government found censorship software on a phone made by Chinese company Xiaomi. It censored content and sent data to Chinese authorities.
Internet Blackouts
In extreme cases, countries can turn off the entire internet in a region. These internet blackouts force everyone offline and are often imposed during protests or elections.
In 2019, the government in Sudan restricted access to the internet across the entire country. Following a court case that found the restrictions were unlawful, access to the internet was restored — but only for the one person who challenged the restrictions in court.
Since 2019, we have recorded more than 500 major government-mandated internet shutdowns. These include internet blackouts, as well as social media shutdowns and severe throttling.
Khartoum, Sudan. 30th June, 2019. Sudanese protest against the ruling Transitional Military Council (TMC) during a total internet blackout.
In total, these restrictions cost the global economy over $44 billion. In 2022, countries including Russia, Myanmar, Iran, India and Ethiopia all implemented internet shutdowns.
Self-Censorship
In many countries, oppressive regimes encourage people to censor themselves, out of fear of reprisals.
In China, for example, journalists, human rights lawyers, and activists have been detained for their online activities. In Iran, journalist Ruhollah Zam, who founded a site critical of the government, was executed. Vietnam has sentenced a number of independent online journalists to prison terms of between two and 10 years.
With penalties such as these, many people are reluctant to share their true thoughts online, and so are unable to form associations with others who hold the same views.
Internet Censorship Around the World
The Freedom on the Net report provides an annual snapshot of government censorship and internet freedoms worldwide.
It ranks countries and considers issues such as barriers to information access, technical filtering of content, and violations of user rights, including surveillance and restrictions on free speech.
Freedom on the Net classifies the online freedom of countries worldwide.
Its latest report found that global internet freedom has declined for 15 years in a row, although 17 countries did show improvements.
Out of 4.5 billion people with internet access:
76% live in countries where people have been arrested or imprisoned for posting content on political, social, or religious issues.
51% live in countries where access to social media platforms was restricted in 2022.
44% live in countries where authorities disconnected internet or mobile networks.
The report’s methodology gives each country a score out of 100 and ranks them. Only four countries – Canada, Costa Rica, Iceland, and Japan – had no internet controls.
Countries with the Most Internet Freedom
Countries with the Least Internet Freedom
1. Iceland
1. China
2. Estonia
2. Myanmar
3. Chile
3. Cuba
4. Canada
4. Iran
5. Costa Rica
5. Russia
6. Germany
6. Vietnam
7. Netherlands
7. Saudi Arabia
8. Japan
8. Pakistan
9. Australia
9. Ethiopia
10. United States
10. Azerbaijan
In this section, we give a brief overview of the state of internet freedom in some of the most highly censored nations:
China
China holds the lowest internet freedom score of any country in the world — a score of 9 out of 100 — and that position has hardened further in recent years.
Provincial-level authorities are now vigorously blocking online content, at times at a scale 10 times larger than the national-level Great Firewall.
China’s censorship apparatus, known as the Golden Shield Project, now operates at multiple layers simultaneously.
In November 2025, the Ministry of State Security issued a formal warning reiterating the illegality of using a VPN for circumvention, signalling a new phase of enforcement pressure on individual users.
The most significant leak in the Great Firewall’s history occurred in September 2025, when over 500GB of source code, work logs, and internal communications from Geedge Networks were exposed.
The documents confirm the DPI engine’s ability to recognize the handshakes used by commercial VPN protocols and block them in real time. They also confirmed the export of this technology to overseas government clients.
Blocked content includes virtually all major international social media platforms, news outlets, search engines, and communication services.
Topics that remain heavily censored domestically include any discussion of the 1989 Tiananmen Square massacre, the Taiwan independence movement, the Falun Gong spiritual practice, LGBT+ issues, and coverage of the Ukraine conflict that diverges from the official Chinese government position.
Russia has dramatically accelerated every aspect of its internet control regime.
Roskomnadzor, the state communications regulator, has blocked Facebook, Instagram, X (formerly Twitter), and all Meta-owned platforms. YouTube has been throttled to near-unusable speeds.
To date, thousands of apps and websites have been blocked in Russia. Roskomnadzor has also restricted access to over 400 VPN services, with Apple removing hundreds of listings from Russia’s App Store following government orders.
Roskomnadzor has also directly targeted VPN protocols: OpenVPN, WireGuard, and IKEv2 were targeted from 2023, and since December 2025, SOCKS5, VLESS, and L2TP have been added to the blocked list.
In July 2025, Russia enacted a law that criminalizes searching for content listed as extremist online — the first time liability has been attached to consumption, rather than distribution, of banned material.
Since May 2025, Russia has been conducting mobile internet shutdowns across the country at an average rate of 2,000 per month.
Despite these heightened restrictions and punitive laws, an estimated 40% of internet users in Russia still use VPNs in order to access the open internet.
Myanmar
Myanmar now ties with China for the worst internet freedom environment in the world.
The junta has imposed sweeping network controls, enacted punitive laws, and now imprisons people for social media activity.
In January 2025, the military regime enacted a cybersecurity law that explicitly restricts the operation of anti-censorship tools in the country, codifying what had previously been de facto practice into formal law.
Myanmar was also identified by Freedom House as one of the countries that has received China’s exported censorship technology.
The regime has gone further than any other government in physically seizing circumvention infrastructure: military forces have begun confiscating Starlink terminals, which had been providing connectivity to aid organisations and pro-democracy resistance groups.
Iran
Iran’s internet freedom score remains one of the lowest in the world. The government continues to operate one of the most restrictive digital environments globally, with most international websites blocked and a history of using complete internet shutdowns to suppress protests.
All major Western social media platforms are blocked, along with most international news sites. Human rights organisations and domestic news sites critical of the government are also blocked.
Western trade sanctions further restrict what services are technically available to Iranian users even if they circumvent blocking.
Iran has escalated its assault on VPNs, introducing new measures to block VPN traffic. VPN usage is extremely common among Iranians, who have historically been among the most active users of circumvention tools globally.
Iranian authorities are now reported to be working toward a broader model of digital isolation that could make VPN circumvention significantly harder than it currently is.
How Is Internet Censorship Documented?
One of the challenges with internet censorship is working out how much it’s happening, and where. When censorship is used to violate human rights, including the right to free expression and the right to peaceful assembly, governments are unlikely to publicize that.
In addition, many of the technologies and techniques used are designed to avoid detection.
If content has been tampered with, most internet users might not notice. Often, they just might see an error message that looks like a network fault.
There are several organizations working to document internet censorship around the world:
Since 2012, the Open Observatory of Network Interference (OONI) has been inviting people to run its software to test the blocking of websites and apps. OONI has collected 1583.7 million measurements which it shares as raw data and contextualizes into research reports. China began blocking OONI in July 2023, which will make it hard to gather reliable test data there.
The Censored Planet Observatory uses automated remote measurement techniques to test access to thousands of websites from vantage points in 221 countries. It complements OONI by bringing additional scale, without the need for volunteers to participate in a measured country.
The Internet Outage Detection Analysis (IODA) tool uses almost real-time monitoring of internet infrastructure to generate data on outages. It’s useful to researchers, but it doesn’t provide context on why there was an outage.
A VPN can bypass some forms of internet censorship. A VPN is ineffective against blackouts that stop all internet communications, but it can help to circumvent blocks imposed on websites, IP addresses, or content hosted in particular countries.
States that apply harsh censorship restrictions may also block VPN traffic, too. However, the most sophisticated VPNs use obfuscation technology to disguise their traffic, which makes it much harder for the authorities to block them.
Which Country Has the Strictest Censorship Laws?
China has the strictest online censorship laws in the world. The ruling communist party has banned criticism of the party or any discussion of the Tiananmen Square massacre of 1989, or the Falun Gong spiritual practice, amongst other topics.
International and independent news sources are also blocked, effectively leaving the government in control of the information most people can access in the country.
