---
title: "Internet Censorship by Country — Which Platforms Are Blocked (September 2026 Index)"
description: "Measured internet censorship by country: which social, messaging, news and VPN platforms are actually blocked, from OONI network measurements for September 2026. Updated monthly. Confirmed blocking and unverified anomalies reported separately, and thin probe coverage labelled rather than hidden."
url: https://vpnrank.io/research/internet-censorship-map
language: en
updated: 2026-10-11
image: https://vpnrank.io/enrich/internet-censorship-map/censorship-two-rankings.png
---

vpnrank.io research

# Internet censorship by country: which platforms are actually blocked

Iran blocked 21 of the 25 platforms in this index in September 2026, more than any other of the 179 countries measured, and Finland and Russia were next with 12 each. Most “most censored countries” lists are assembled from press releases and each other. This one is assembled from network measurements taken inside the countries concerned — 3,444 country-by-platform cells covering 25 platforms across 179 countries, for September 2026. Where the measurements are thin, it says so instead of printing a zero.

Updated monthly Current edition: **September 2026** · previous: August 2026 · next update 5 November 2026

Measurement window

September 2026 (`2026-09-01` to `2026-10-01`, end exclusive)

Data source

[OONI](https://ooni.org/) — Open Observatory of Network Interference

Data pulled

2026-10-11. Re-pulled monthly; 3 editions so far.

Analysis

[Sofía Giménez](https://vpnrank.io/authors/sofia-gimenez), Privacy, Jurisdiction & No-Logs Reviewer

## Which country censors the internet the most?

Iran blocks the most platforms in vpnrank.io’s index: 21 of the 25 measured were confirmed blocked at national scale in September 2026, all of them on most networks measured. Finland and Russia were next with 12 each, then Myanmar (2), India (1), Thailand (1) and Türkiye (1). China confirmed no national-scale block but returned elevated anomaly rates on 21 of the 25 platforms, which the index reports as possible interference, never as confirmed blocks.

## Which countries block VPN websites?

In September 2026, Iran blocked all 8 VPN provider websites we could measure there (90.3–96.9% of measurements), Russia 7 of 8 (18.6–26.9%) and Finland 7 of 8 (3.2–9.2%). No other country confirmed a national-scale block on one, apart from 1 country flagged as possibly measured on a locally filtered network (a school, office or campus connection) and not counted. A blocked website is not a blocked service: this shows the sign-up and download pages are unreachable, not that an installed VPN fails to connect.

## Which countries block social media?

In September 2026, 6 countries confirmed blocking a social media or messaging platform at national scale: Iran (all 9 we measure), Finland (Facebook, Instagram and Twitter), Russia (Facebook, Instagram and Twitter), Myanmar (Instagram and Twitter), India (TikTok) and Türkiye (Discord). In every other country these platforms were reachable, blocked only on isolated networks, flagged with unconfirmed anomalies or too thinly measured to judge.

“Confirmed at national scale” means OONI confirmed the block in at least 2% of a country’s measurements of that platform. Anomalies and countries flagged as possibly measured on a locally filtered network are never counted. OONI data for September 2026, pulled 2026-10-11; the matrix, the thresholds and every caveat are below.

The three rules this page is built on

-   **Confirmed blocking and anomalies are never added together.** They are different claims with different confidence, and merging them is the single most common way a censorship figure becomes wrong.
-   **A confirmed block is always divided by measurements before it means anything.** Four blocks out of fourteen thousand is not the same event as three thousand out of five thousand.
-   **Thin coverage is reported as thin coverage.** A country nobody measured is not a country without censorship, and it never appears here as an empty cell that implies access.

Key findings, September 2026

-   **Iran blocks 21 of the 25 platforms measured**, on most networks measured, across every category in the test list — the only country in this edition where the picture needs no hedging.
-   **All 8 VPN provider websites we could measure are confirmed blocked in Iran**, at 90.3% to 96.9% of measurements. You would already need a VPN to reach the page where you buy one.
-   **China shows 21 platforms flagged as anomalies and almost no confirmed blocks.** A confirmed-only ranking puts it 20th of 179, below the United Kingdom, Canada and France — which is why this index never ranks on confirmed counts.
-   **Russia blocks 12 platforms, but at 6.3–26.9% of measurements rather than above 80%.** That is a network-by-network regime: a Russian user’s experience depends on their ISP.
-   **1,285 of 3,444 cells (37%) have too little data to say anything**, and are labelled as such rather than shown as accessible.
-   **All 12 newly confirmed blocks since August 2026 are in Finland, and every one is on a VPN or Tor domain.** That is a filtered-network artefact rather than censorship, and it is excluded from the ranking below.

Free to quote with attribution to vpnrank.io and to [OONI](https://ooni.org/) as the data source. Please state the measurement window — a censorship figure without its window is not a fact.

## What changed since August 2026

The largest movement in this edition is not censorship. Between August 2026 and September 2026, 12 cells crossed from no confirmed blocking into confirmed blocking — and 8 of them are VPN-provider or Tor domains in Finland, whose social, news and reference platforms stayed clean throughout. That is the signature of a content filter with an anonymiser category switched on appearing in the probe pool, not of a new blocklist.

We report it as a change because it *is* one — the measurements really did move — and then exclude those countries from the ranking, which is what the flag is for. An index that quietly dropped the rows would be easier to read and impossible to check.

| Country | Platform | Category | August 2026 | September 2026 | What it doesn't automatically prove |
| --- | --- | --- | --- | --- | --- |
| Finland | Facebook | Social platforms | 0.0% confirmed | 5.2% confirmed (n=3,123) | A month-over-month move can reflect a change in which networks were measured as much as a change in policy. The rate and the measurement count are both shown so the reader can judge. |
| Finland | Instagram | Social platforms | 0.0% confirmed | 5.2% confirmed (n=3,131) | A month-over-month move can reflect a change in which networks were measured as much as a change in policy. The rate and the measurement count are both shown so the reader can judge. |
| Finland | Twitter (twitter.com) | Social platforms | 0.0% confirmed | 5.2% confirmed (n=3,130) | A month-over-month move can reflect a change in which networks were measured as much as a change in policy. The rate and the measurement count are both shown so the reader can judge. |
| Finland | BBC | News and video | 0.0% confirmed | 7.2% confirmed (n=152) | A month-over-month move can reflect a change in which networks were measured as much as a change in policy. The rate and the measurement count are both shown so the reader can judge. |
| Finland | Tor Project | Circumvention tools | 0.0% confirmed | 8.6% confirmed (n=116) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | ExpressVPN | VPN provider sites | 0.0% confirmed | 7.0% confirmed (n=172) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | NordVPN | VPN provider sites | 0.0% confirmed | 4.6% confirmed (n=239) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | Surfshark | VPN provider sites | 0.0% confirmed | 3.2% confirmed (n=189) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | Proton VPN | VPN provider sites | 0.0% confirmed | 4.0% confirmed (n=224) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | PureVPN | VPN provider sites | 0.0% confirmed | 9.2% confirmed (n=119) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | CyberGhost | VPN provider sites | 0.0% confirmed | 6.2% confirmed (n=178) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Finland | Private Internet Access | VPN provider sites | 0.0% confirmed | 9.1% confirmed (n=121) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | Tor Project | Circumvention tools | 19.2% confirmed | 26.7% confirmed (n=1,130) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | ExpressVPN | VPN provider sites | 18.7% confirmed | 25.2% confirmed (n=1,254) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | Surfshark | VPN provider sites | 18.1% confirmed | 22.3% confirmed (n=1,429) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | Proton VPN | VPN provider sites | 18.5% confirmed | 21.5% confirmed (n=1,597) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | CyberGhost | VPN provider sites | 18.6% confirmed | 24.3% confirmed (n=1,255) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |
| Russia | Private Internet Access | VPN provider sites | 19.9% confirmed | 26.4% confirmed (n=1,144) | Blocking confined to circumvention domains, with everything else clean, is what a corporate or campus filter looks like from inside the probe pool. It is not evidence of a state blocklist. |

Cells that moved into or further up the confirmed-blocking scale between August 2026 and September 2026. OONI data, pulled 2026-10-11.

In the other direction, 14 cells dropped out of confirmed blocking. Most fell below the 2% threshold rather than reaching zero, which is a smaller claim than “unblocked” and is stated that way.

| Country | Platform | August 2026 | September 2026 | What it doesn't automatically prove |
| --- | --- | --- | --- | --- |
| Germany | Tor Project | 3.0% confirmed | 0.2% confirmed (n=457) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Mexico | Tor Project | 3.5% confirmed | 0.0% confirmed (n=67) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | ExpressVPN | 3.6% confirmed | 0.2% confirmed (n=545) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Mexico | ExpressVPN | 3.9% confirmed | 0.0% confirmed (n=96) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Türkiye | ExpressVPN | 2.1% confirmed | no confirms, but anomalies still elevated (n=92) | Confirmed blocking stopping while anomalies stay high is not access restored. It can equally mean the blocking method changed to one OONI cannot confirm. |
| Germany | NordVPN | 2.5% confirmed | 0.2% confirmed (n=892) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | Surfshark | 6.2% confirmed | 0.1% confirmed (n=690) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | Proton VPN | 3.1% confirmed | 0.1% confirmed (n=713) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | PureVPN | 3.3% confirmed | 0.2% confirmed (n=432) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Mexico | PureVPN | 3.3% confirmed | 1.4% confirmed (n=73) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | CyberGhost | 3.6% confirmed | 0.4% confirmed (n=558) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Mexico | CyberGhost | 3.2% confirmed | 1.0% confirmed (n=98) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Germany | Private Internet Access | 3.4% confirmed | 0.2% confirmed (n=443) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |
| Mexico | Private Internet Access | 3.4% confirmed | 1.4% confirmed (n=72) | Falling below the threshold means the block was seen on fewer networks this month, not that it was removed. |

Cells that left the confirmed-blocking scale between August 2026 and September 2026. OONI data, pulled 2026-10-11.

**The signal worth watching this month is anomalies on social platforms.** 12 appeared. Puerto Rico accounts for 5 of them, arriving together on Facebook, Instagram, YouTube, BBC, The New York Times; the remaining 4 are isolated single platforms in Oman, Myanmar (Burma), Syria, Ethiopia, which on their own mean little. Anomalies are not a blocking claim and never enter the ranking. But several arriving at once on social platforms in one country is the shape an actual event makes, and it is the thing to check against the next edition.

Measurement coverage also moved: 38 cells gained enough measurements to be judged and 206 lost it. A country that becomes measurable — or stops being measurable — changes what this index can say about it without anything changing on its networks.

## Counting confirmed blocks produces a censorship ranking topped by Germany and the United States

The obvious way to build this page is to ask OONI how many measurements in each country came back “confirmed blocked”, add them up, and sort. Doing exactly that on this month’s data puts Germany fifth with 72 confirmed blocks, the United States sixth and the United Kingdom eleventh, and it leaves China in eighteenth place — below Canada, Switzerland and France.

The reason is that a confirmed count is an absolute number measured over wildly different volumes. The United States contributes 630,740 measurements in this window and the United Kingdom 139,768; Iran contributes 40,518. A large, heavily-measured country accumulates a long tail of confirmed blocks from school networks, corporate filters and parental controls, and that tail is arithmetically larger than the entire measured output of a country that actually censors.

| Country | Confirmed blocks | Measurements | Confirmed rate | What the raw count doesn't automatically prove |
| --- | --- | --- | --- | --- |
| Iran | 28,205 | 40,518 | 69.61% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| Russia | 12,955 | 388,083 | 3.34% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| India | 2,612 | 82,557 | 3.16% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| Türkiye | 2,017 | 37,997 | 5.31% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| United States | 589 | 630,740 | 0.09% | That a country appears high here says more about how much it was measured than about what it blocks. At this rate the confirmed blocks are individual filtered networks. |
| Finland | 580 | 25,693 | 2.26% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| Myanmar (Burma) | 252 | 8,789 | 2.87% | A rate this high across multiple categories is consistent with national blocking, but the count alone was never what established that — the rate was. |
| United Kingdom | 235 | 139,768 | 0.17% | That a country appears high here says more about how much it was measured than about what it blocks. At this rate the confirmed blocks are individual filtered networks. |

Raw confirmed blocks summed across all 25 platforms, August 2026, with the confirmed rate that the raw count hides. Derived by vpnrank.io from OONI aggregation data pulled 2026-10-11.

![Side-by-side rankings of the same OONI August 2026 measurements: ranked by raw confirmed blocks, Iran, Russia, India, Türkiye and Germany lead with the United States sixth and China eighteenth; ranked by platforms blocked above 2% of measurements, Iran has 21, Russia 12, and Germany and Mexico are flagged and excluded](https://vpnrank.io/enrich/internet-censorship-map/censorship-two-rankings.png)

The same measurements, read two ways. Verified September 2026.

So every country figure on this page is a **rate**. A platform counts as blocked at national scale in a country when confirmed blocks exceed 2% of that country’s measurements for it, and as widespread above 20%. Below the lower threshold, the block is real but local — one ISP, one school, one company — and it is excluded from every ranking here.

> Practical ruleAny censorship statistic that reports a count rather than a rate is measuring probe volume, not censorship. Ask what the denominator was before you cite the number.

## China’s blocking is invisible to a confirmed-only view

There is a second trap underneath the first, and it is worse because correcting the first does not fix it. OONI can only mark a measurement “confirmed blocked” when it observes positive evidence of blocking, such as a block page served in place of the real site. Blocking that works by DNS tampering or by resetting the connection produces an *anomaly* instead — a strong signal that something interfered, without the artefact that would confirm it.

China is the clearest case in the dataset. Across the 25 platforms measured, 21 come back with elevated anomaly rates and essentially none come back confirmed. A view built on confirmed blocking alone therefore reports China as freer than the United Kingdom, which is a conclusion no reader would accept and no honest method should produce.

| Platform | Category | Anomaly rate | Confirmed blocks | Measurements | What this doesn't automatically prove |
| --- | --- | --- | --- | --- | --- |
| Wikipedia (EN) | Reference and search | 97.4% | 0 | 39 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| X (x.com) | Social platforms | 78.6% | 0 | 234 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Google Search | Reference and search | 78.4% | 0 | 51 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| The New York Times | News and video | 66.2% | 0 | 71 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Surfshark | VPN provider sites | 60.2% | 0 | 98 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| ExpressVPN | VPN provider sites | 59.5% | 0 | 79 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Tor Project | Circumvention tools | 55.4% | 0 | 65 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Reddit | Social platforms | 51.9% | 0 | 208 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Telegram | Messaging | 49.5% | 1 | 212 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Signal | Messaging | 46.1% | 1 | 204 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| TikTok | Social platforms | 44.9% | 0 | 245 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |
| Discord | Messaging | 43.6% | 3 | 234 | An anomaly is a sign of possible interference with documented false positives. It is a flag for investigation, not a blocking claim, and it is never added to the confirmed figures. |

The twelve highest anomaly rates recorded in China, August 2026, out of 21 platforms flagged. OONI data, pulled 2026-10-11.

One row in the China data deserves its own caveat: Netflix comes back as accessible on 45 measurements. That is above our 30\-measurement floor but still thin, and more importantly it measures whether the domain is reachable — not whether the service will serve a viewer there. Network reachability and service availability are different questions, and this dataset only answers the first.

## Iran blocks 21 of 25 platforms — including every VPN website we could measure

Iran has its own page in this index, with every platform, its rate and the month-over-month change: [Internet censorship in Iran (September 2026)](https://vpnrank.io/research/internet-censorship-map/iran).

Iran is the one country in this window where the measured picture needs no hedging. 21 of the 25 platforms are confirmed blocked on most networks measured, spanning every category in the test list: social, messaging, news and video, circumvention tools, and the websites of VPN providers themselves. The rates are not marginal — Facebook is confirmed blocked in 77.3% of Iranian measurements.

The VPN-provider category is the reason we built the target list ourselves rather than reusing a standard one. These are the nine providers in our own database — the set we recommend from — and we measured their public websites the same way we measured everything else. In Iran, 8 of them are confirmed blocked at rates between 90.3% and 96.9%.

![Bar chart of the share of OONI measurements in which each VPN provider's own website was confirmed blocked in August 2026: in Iran between 81.7% and 87.1% across eight providers, and in Russia between 15.3% and 20.0%](https://vpnrank.io/enrich/internet-censorship-map/vpn-site-blocking.png)

The websites of the providers we recommend, measured against the same test as every other platform. Verified September 2026.

| Provider website | Iran — confirmed rate | Iran — measurements | Russia — confirmed rate | What it doesn't automatically prove |
| --- | --- | --- | --- | --- |
| IPVanish | 96.9% | 192 | No confirmed blocking above threshold | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| Private Internet Access | 95.6% | 181 | 26.4% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| CyberGhost | 95.3% | 211 | 24.3% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| ExpressVPN | 94.0% | 200 | 25.2% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| NordVPN | 93.0% | 213 | 18.6% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| PureVPN | 92.3% | 183 | 26.9% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| Surfshark | 90.5% | 211 | 22.3% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |
| Proton VPN | 90.3% | 216 | 21.5% | A blocked website is not a blocked service. This measures whether the sign-up, payment and download pages load — not whether an already-installed application connects. |

VPN provider websites measured in Iran and Russia, August 2026. TotalVPN returned no measurements at all — OONI's test lists do not appear to cover the domain, so it is reported as unmeasured and must not be read as unblocked. OONI data, pulled 2026-10-11. vpnrank.io earns affiliate commission from some of the providers named.

The practical reading is narrow and worth stating precisely, because the tempting version is wrong. This does not show that a VPN fails to work in Iran; it shows that the pages you would use to *obtain* one are unreachable from most measured Iranian networks. The consequence is about sequence rather than capability — the moment to install and pay for a tool is before you are behind the border that blocks its shop.

> Practical ruleA country that blocks circumvention websites has made the download, not the connection, the scarce step. Judge a tool by whether it can be acquired in advance and used offline-first, not only by whether it connects.

## Russia’s blocking is partial by design, and that is a finding rather than a caveat

Russia sits second on the measured ranking with 12 platforms blocked at national scale, but the shape of its numbers is completely different from Iran’s. Where Iranian confirmed rates cluster above 80%, Russian rates sit at 6.3–26.9% — real, repeated, confirmed blocking that reaches a minority of the networks measured rather than nearly all of them.

That is what a network-by-network blocking regime looks like in measurement data: some ISPs implement an order and others have not, or implement it differently. Reporting it as “Russia blocks Facebook” without the rate would be true in spirit and misleading in practice, because the measured experience of a Russian user depends on which provider they are on. Four further platforms — X (x.com), Telegram, Discord, YouTube — show elevated anomalies without confirmation, and are listed separately below rather than folded into the count.

## Germany and Mexico look like censors in the raw output, and are not

Two countries rank third and fourth on the measured scale with 0 platforms each above the national-scale threshold: Germany and Mexico. Both are flagged by the collector and excluded from the ranking, and the reason is visible in the category breakdown rather than in the totals.

| Country | Platforms above threshold | Categories affected | Social, news and reference | What the flag means |
| --- | --- | --- | --- | --- |
| Mexico | 3 | NordVPN, Surfshark, Proton VPN | No confirmed blocking above the threshold on any social, news or reference platform | Confirms restricted entirely to VPN and Tor domains are the fingerprint of a content filter with an anonymiser category enabled — a filtered network inside the probe pool, not a state blocklist. The flag excludes the country from the ranking; it does not by itself clear or accuse it. |

Countries flagged as filtered-network suspects in the August 2026 pull, with the evidence behind the flag. Both were checked by hand before publication. OONI data, pulled 2026-10-11.

The confirmed rates make the same point quantitatively: in Germany the VPN-provider domains sit between roughly 2.5% and 6.2% of measurements, and every social, messaging and news platform in the list is either accessible or shows only isolated confirms — 16 of them. A state blocklist does not look like that. A corporate or campus web filter with an “anonymisers” category switched on looks exactly like that, and such networks are well represented among volunteer OONI probes.

We publish the flag rather than silently dropping the rows, because the alternative — a censorship map that quietly decides which inconvenient countries to omit — is not inspectable. A reader who disagrees with our reading can see the same numbers and reach their own.

## The noisiest signal in the dataset is anomalies on VPN and Tor websites

If anomalies were a reliable blocking signal, the list of countries showing them on circumvention domains would look like a list of authoritarian states. It does not. In this window, Switzerland (9), Kazakhstan (7), Denmark (6), Kenya (4) all show elevated anomaly rates *only* on VPN-provider and Tor domains, with no confirmed national-scale blocking anywhere in the list. Some of that is filtered networks in the probe pool again; some is plausibly the domains themselves, which sit behind aggressive anti-abuse infrastructure that can produce connection failures indistinguishable from interference at measurement level.

The contrast is what makes the flag useful. When elevated anomalies appear on *social and news* platforms instead — which they do in Tanzania (5), Cambodia (4) — the ordinary explanations stop fitting, and the case for looking harder gets stronger. Neither pattern is a blocking claim. One is worth an investigation and the other mostly is not.

## The country-by-platform matrix

Below is every country in the pull that showed any signal at all — 29 of 179 — against all 25 platforms measured. Countries not listed showed no confirmed blocking above the isolated threshold and no elevated anomalies; that is a measured result for the ones with coverage and an absence of evidence for the ones without, which the next section separates out.

BConfirmed, widespread PConfirmed, partial AElevated anomalies iIsolated networks ·Accessible ?Not measured

| Country | Facebook | Instagram | Twitter (twitter.com) | X (x.com) | TikTok | Reddit | Telegram | Signal | Discord | YouTube | BBC | The New York Times | Netflix | Wikipedia (EN) | Google Search | Tor Project | ExpressVPN | NordVPN | Surfshark | Proton VPN | PureVPN | CyberGhost | Private Internet Access | IPVanish | TotalVPN |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Iran | B | B | B | B | B | B | B | B | B | B | B | · | B | · | · | B | B | B | B | B | B | B | B | B | ? |
| Russia | P | P | P | A | · | · | A | · | A | A | P | · | · | · | · | B | B | P | B | B | B | B | B | · | ? |
| Finland | P | P | P | · | · | · | · | · | · | · | P | · | · | · | · | P | P | P | P | P | P | P | P | · | ? |
| Mexico(flagged) | i | i | i | · | · | · | · | · | · | · | i | · | · | ? | · | · | · | P | P | P | i | i | i | · | ? |
| Myanmar (Burma) | A | P | P | A | · | · | · | A | A | · | · | · | · | · | · | A | A | A | A | A | A | A | · | A | ? |
| Türkiye | i | i | i | · | · | · | · | · | B | · | i | · | · | · | · | · | A | A | A | A | A | A | A | A | ? |
| India | i | i | i | · | B | · | · | · | · | · | i | · | · | · | · | i | i | i | i | i | i | i | i | · | ? |
| Thailand | · | · | · | · | · | · | · | · | · | · | · | · | ? | P | ? | · | · | · | · | · | · | · | · | · | ? |
| China | A | A | A | A | A | A | A | A | A | A | A | A | · | A | A | A | A | A | A | i | A | · | A | A | ? |
| Puerto Rico | A | A | · | · | · | · | · | · | · | A | A | A | ? | ? | ? | A | A | A | A | A | · | · | A | A | ? |
| Venezuela | · | · | · | · | · | · | · | A | · | · | · | · | · | · | · | A | A | A | A | A | A | A | A | A | ? |
| Switzerland | i | i | i | · | · | · | · | · | · | · | i | · | · | · | · | A | A | A | A | A | A | A | A | A | ? |
| Kazakhstan | · | · | · | · | · | · | · | · | · | · | · | · | ? | ? | ? | ? | A | A | A | A | ? | A | A | A | ? |
| Denmark | · | · | · | · | · | · | · | · | · | · | ? | · | ? | ? | ? | A | ? | A | A | A | A | ? | A | ? | ? |
| Pakistan | · | · | · | · | · | · | A | · | · | · | · | · | ? | · | ? | ? | A | A | · | A | ? | A | ? | ? | ? |
| Tanzania | · | · | A | A | · | · | A | A | A | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Belarus | · | · | · | · | · | · | · | · | · | · | A | · | · | · | · | A | · | · | A | A | · | · | · | · | ? |
| Kenya | · | · | · | · | · | · | · | · | · | · | · | · | · | · | · | · | · | · | · | A | A | A | A | · | ? |
| Cambodia | A | A | A | · | · | · | · | · | A | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Uzbekistan | · | · | · | · | A | · | · | · | · | · | · | ? | ? | ? | ? | ? | ? | A | · | · | ? | ? | ? | ? | ? |
| Afghanistan | · | · | · | · | A | · | · | · | · | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Ethiopia | · | · | · | · | · | · | · | · | · | · | A | · | ? | ? | ? | · | · | · | · | · | · | · | · | · | ? |
| Iceland | · | · | · | · | · | · | · | A | · | · | · | ? | ? | ? | ? | ? | ? | · | · | · | ? | · | ? | ? | ? |
| Jordan | · | · | · | · | A | · | · | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Kyrgyzstan | · | · | · | · | A | · | · | · | · | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Oman | · | · | · | · | A | · | · | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Syria | · | · | · | · | · | · | · | · | A | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Uganda | A | · | · | · | · | · | · | · | · | · | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? | ? |
| Vietnam | · | · | · | · | · | · | · | · | · | · | A | · | · | ? | · | · | · | · | · | · | · | · | · | · | ? |

Country × platform status for August 2026. B = confirmed on at least 20% of measurements; P = confirmed on 2–20%; A = anomaly rate at or above 20% with no confirmation, which is a flag and not a blocking claim; i = confirmed on under 2% of measurements, i.e. individual filtered networks; ? = fewer than 30 measurements, which says nothing in either direction. Data: [OONI](https://ooni.org/), pulled 2026-10-11.

## A third of the grid was never measured, and pretending otherwise would be the real error

Of the 3,444 country-by-platform cells in this pull, 1,285 — 37% — fall below the 30\-measurement floor. OONI runs on volunteer probes, so coverage follows where volunteers are, not where censorship is. 49 countries have usable coverage on all 25 platforms; others have almost none.

| Country | Platforms with too little data | Platforms with usable coverage | What this doesn't automatically prove |
| --- | --- | --- | --- |
| Lesotho | 23 of 25 | 2 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Papua New Guinea | 23 of 25 | 2 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Liberia | 21 of 25 | 4 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Timor-Leste | 19 of 25 | 6 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| United Arab Emirates | 14 of 25 | 11 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Bosnia & Herzegovina | 14 of 25 | 11 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Bangladesh | 14 of 25 | 11 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |
| Bolivia | 14 of 25 | 11 | Nothing about whether these platforms are blocked. Thin coverage is an absence of evidence, and reporting it as an absence of censorship is how a measurement dataset gets turned into a false all-clear. |

The eight countries with the least usable coverage in the August 2026 pull. Cells below 30 measurements are excluded from every figure on this page rather than counted as accessible. OONI data, pulled 2026-10-11.

This is also the reason the matrix above uses a distinct mark for unmeasured cells rather than leaving them blank. A blank cell in a censorship table reads as “fine here”, and for roughly a third of this grid that reading would be unsupported.

## Method, in enough detail to be checked

Every figure here comes from OONI’s aggregation API for the `web_connectivity` test, one request per target domain returning all countries at once, for the window 2026-09-01 to 2026-10-01 (end exclusive). The target list is 25 domains across six categories, and the exact hostname for each was probed individually — `nordvpn.com` carries measurements where `www.nordvpn.com` carries none, and for other providers it is the other way round, so a “tidied” hostname would silently produce a zero that reads as “nowhere blocked”.

| Status | Rule | What it licenses you to say | What it doesn't automatically prove |
| --- | --- | --- | --- |
| Confirmed, widespread | Confirmed blocks ≥ 20% of measurements | That the country blocks the platform on most networks measured. | Not that every network blocks it, and not that the block is legally mandated rather than operator-led. |
| Confirmed, partial | Confirmed blocks 2–20% of measurements | That real blocking exists on a meaningful minority of networks — typically some ISPs and not others. | Not blanket national blocking. A user's experience depends on their provider. |
| Elevated anomalies | ≥ 30 measurements and anomaly rate ≥ 20% | That something interfered often enough to be worth investigating. | Nothing about blocking. Anomalies have documented false positives and are never counted as blocks here. |
| Isolated networks | Confirmed blocks > 0 but < 2% | That at least one network blocks it — a school, an employer, a filter, one ISP. | Not that the country blocks anything. These are excluded from every ranking on this page. |
| Accessible | ≥ 30 measurements, no confirmed blocks, anomaly rate < 20% | That it was reachable in the measurements we have. | Not that the service works, is legal, or will serve a visitor — only that the domain responded normally. |
| Not measured | < 30 measurements | Nothing. | Above all, not access. This status exists precisely so that thin coverage cannot be mistaken for an all-clear. |

The six statuses and the thresholds behind them, written into the published data file so any figure here can be audited against it. Thresholds calibrated on the August 2026 distribution. Verified September 2026.

**Repeatability:** the collector is a script in our repository, the thresholds are constants inside it, and every threshold is written into the output file alongside the results. Re-running it against a later month regenerates this page in full. It is pulled monthly.

**Limitations we are aware of.** OONI probes are volunteer-run, so coverage is uneven and skews toward networks where volunteers install the app — which includes filtered corporate and campus networks, the source of the Germany and Mexico artefact. Measurements are of domain reachability, not of service usability. A month-long window smooths over short blocking events, which will under-report protest-driven shutdowns. And the target list is 25 platforms chosen by us; a different list would produce different country totals, which is why the totals are always reported as “of 25 platforms measured” rather than as an absolute score.

## Source, licence and how to cite this

The measurements are collected and published by [OONI, the Open Observatory of Network Interference](https://ooni.org/), a free software project that has been running network measurement tests through volunteer probes since 2012. OONI publishes its data openly under a Creative Commons licence; the analysis, thresholds, category choices and every judgement about what the numbers do and do not support are ours, and OONI has not reviewed or endorsed them.

Reuse and citation

This analysis is free to reuse for journalism and research with attribution. Please credit OONI as the data source alongside us, and state the measurement window — a censorship figure without its window is not a fact.

Citation: vpnrank.io. *Internet censorship by country: which platforms are actually blocked*. Analysis of OONI web connectivity data, September 2026 edition. Accessed \[date\]. https://vpnrank.io/research/internet-censorship-map

This URL always holds the current edition, so a link to it stays live as the data is re-pulled. If you need the edition you originally cited, ask and we will send that month’s file — every pull is kept.

[Request the underlying data cut or a specific country →](https://vpnrank.io/press)

How we are funded, and why it is on this page

vpnrank.io earns affiliate commission when readers subscribe to some of the providers we cover, including providers named in the VPN-website section above. That is exactly why the measurement for that section was run identically to every other category, why the thresholds were fixed before the results were seen, and why the provider that returned no data is reported as unmeasured rather than dropped. You can check the arithmetic against OONI’s own API; that is the point of publishing the method.

Related research and tools: [the VPN Price Index](https://vpnrank.io/vpn-price-index) tracks what these providers charge over time, [our China guide](https://vpnrank.io/best-vpn/china) and [Iran guide](https://vpnrank.io/best-vpn/travel/iran) cover the two countries above in practical detail, and the [research hub](https://vpnrank.io/research) lists everything else we publish with a documented method.

Structured data (JSON-LD):

```json
{"@context":"https://schema.org","@type":"Organization","@id":"https://vpnrank.io/#organization","name":"VPNRank","alternateName":"vpnrank.io","url":"https://vpnrank.io","email":"hello@vpnrank.io","logo":{"@type":"ImageObject","url":"https://vpnrank.io/logos/favicon.png","width":500,"height":500},"description":"Independent VPN comparison. We track 9 major VPN services every day and rank them for privacy, security, speed and value.","foundingDate":"2026-03","sameAs":["https://www.youtube.com/@vpnrank","https://www.instagram.com/vpnrank.io","https://www.facebook.com/people/VPNRankio/61592338017872/"],"knowsAbout":["Virtual Private Networks","Internet Privacy","Cybersecurity","Network Encryption","VPN Protocols"]}
```

```json
{"@context":"https://schema.org","@type":"Organization","@id":"https://vpnrank.io/#editorial-team","name":"VPNRank Editorial Team","url":"https://vpnrank.io/about","parentOrganization":{"@id":"https://vpnrank.io/#organization"},"description":"A team of privacy and cybersecurity analysts who independently test VPN services for speed, security, leak protection, and value.","knowsAbout":["VPN testing methodology","WireGuard and OpenVPN protocols","DNS and IPv6 leak testing","No-logs policy audits","Encryption standards"]}
```

```json
{"@context":"https://schema.org","@type":"WebSite","@id":"https://vpnrank.io/#website","name":"VPNRank","alternateName":"vpnrank.io","url":"https://vpnrank.io","publisher":{"@id":"https://vpnrank.io/#organization"},"inLanguage":"en-US"}
```

```json
{"@context":"https://schema.org","@type":"Article","headline":"Internet censorship by country: which platforms are actually blocked (September 2026)","description":"Measured internet censorship by country: which social, messaging, news and VPN platforms are actually blocked, from OONI network measurements for September 2026. Updated monthly. Confirmed blocking and unverified anomalies reported separately, and thin probe coverage labelled rather than hidden.","datePublished":"2026-09-02","dateModified":"2026-10-11","url":"https://vpnrank.io/research/internet-censorship-map","mainEntityOfPage":{"@type":"WebPage","@id":"https://vpnrank.io/research/internet-censorship-map"},"image":"https://vpnrank.io/enrich/internet-censorship-map/censorship-two-rankings.png","author":{"@type":"Person","name":"Sofía Giménez","url":"https://vpnrank.io/authors/sofia-gimenez","jobTitle":"Privacy, Jurisdiction & No-Logs Reviewer"},"publisher":{"@type":"Organization","name":"vpnrank.io","url":"https://vpnrank.io"},"isBasedOn":{"@type":"Dataset","name":"OONI web connectivity measurements","url":"https://ooni.org/data/","creator":{"@type":"Organization","name":"Open Observatory of Network Interference (OONI)","url":"https://ooni.org/"}}}
```

```json
{"@context":"https://schema.org","@type":"Dataset","name":"Internet censorship by country and platform — September 2026 edition","description":"Country-by-platform internet blocking status for 25 platforms across 179 countries, derived from OONI web connectivity measurements. Confirmed blocking and unverified anomalies are reported separately and never summed.","url":"https://vpnrank.io/research/internet-censorship-map","identifier":"https://vpnrank.io/research/internet-censorship-map","datePublished":"2026-09-02","dateModified":"2026-10-11","temporalCoverage":"2026-09-01/2026-10-01","isAccessibleForFree":true,"creator":{"@type":"Organization","name":"vpnrank.io","url":"https://vpnrank.io"},"isBasedOn":{"@type":"Dataset","name":"OONI web connectivity measurements","url":"https://ooni.org/data/","creator":{"@type":"Organization","name":"Open Observatory of Network Interference (OONI)","url":"https://ooni.org/"}},"variableMeasured":[{"@type":"PropertyValue","name":"confirmed blocking rate","description":"Confirmed blocked measurements divided by total measurements for a country and platform."},{"@type":"PropertyValue","name":"anomaly rate","description":"Measurements flagged anomalous divided by total. A sign of possible interference with documented false positives — never a blocking claim."},{"@type":"PropertyValue","name":"measurement count","description":"Total measurements. Below 30 a cell is reported as not measured, never as accessible."}]}
```