WorldRisk

Which passport do you hold?

Risk to a traveller depends on their nationality. A country that sanctions your destination, or is sanctioned by it, changes how you are treated on arrival. Choosing here adjusts every score on the site.

We hold travel advice written for four nationalities so far. Selecting anything else shows the neutral baseline, and every page says so where it applies rather than presenting it as an assessment made for you.

Kept in your browser only. Nothing is sent anywhere, and you can change it at any time from the header.

All countries

Burundi

66High66 of 100, High band, 50 to 74moderate confidence(6/7 dimensions)
What this score means
Extreme75 to 100
Severe and systematic. Device search, interception and detention over online activity are documented rather than theoretical. Reconsider what you carry and what you say.
High50 to 74
Substantial and consistently evidenced measures. Device inspection, monitored networks or restricted services should be expected rather than treated as unlikely.
Caution25 to 49
Some risk, and manageable with precautions. Measures are evidenced but are narrow, inconsistently applied, or the picture is mixed.
Low0 to 24
Ordinary precautions. No systematic device search, network restriction or prosecution of visitors for online speech is evidenced.

Government travel advice is written for one citizenry, so the same country carries a different assessment depending on your passport. Change it in the header.

Showing the baseline, which weights every issuer equally. No government in our corpus publishes travel advice written for you, so this is not addressed to you specifically.

Dimensions

Each dimension is scored independently, because the traveller who cares about legal exposure for what they have posted is often not the one carrying client data. Higher means more risk.

Speech and legal exposure

What legal risk attaches to what I have said online, or say while I am there?

86Extreme86 of 100, Extreme band, 75 to 100

good confidence8 independent sources

Why this score (11 sources)

  • Freedom House 2026 scored BDI 0 of 4 on freedom to express personal views on sensitive topics without fear of surveillance or retribution.

    Freedom House, Freedom in the World · observed 2026-01-01

  • Freedom House 2026 scored BDI 0 of 4 on free and independent media.

    Freedom House, Freedom in the World · observed 2026-01-01

  • V-Dem Digital Society Project 2025 scored BDI 2.60 of 4 on how routinely the government censors political content on social media, where higher is freer.

    V-Dem Institute, University of Gothenburg, Digital Society Project · observed 2025-12-31

  • V-Dem Digital Society Project 2025 scored BDI 0.63 of 4 on how often people are arrested for political content they posted online, where higher is freer.

    V-Dem Institute, University of Gothenburg, Digital Society Project · observed 2025-12-31

  • Burundi - Level 3: Reconsider Travel

    US Department of State, Travel Advisories · observed 2025-04-29 · written for USA nationals

  • FCDO advice for Burundi: avoid_all_travel_to_parts.

    UK Foreign, Commonwealth and Development Office, Foreign travel advice · observed 2026-05-21 · written for GBR nationals

  • Burundi: Avoid non-essential travel

    Global Affairs Canada, Travel advice and advisories · observed 2026-09-24 · written for CAN nationals

  • Burundi: No travel warning in force

    Auswärtiges Amt, Reise- und Sicherheitshinweise · observed 2026-09-01 · written for DEU nationals

  • In December 2024, Sandra Muhoza, a journalist from the online media outlet La Nova Burundi , was sentenced to 21 months in prison for “attacking the integrity of the national territory and racial aversion.”

    Global Voices (Advox), via a site scoped search · observed 2025-04-25 · extracted by language model

  • Comme nous l′avons signalé, la Haute Cour de Mukaza l′a déclarée coupable d′atteinte à l′intégrité du territoire national et d′incitation à la haine ethnique en décembre 2024, pour des commentaires dans un groupe WhatsApp de journalistes, et l′a condamnée à 21 mois de prison.

    Our translation: As we reported, the Mukaza High Court found her guilty of undermining the integrity of the national territory and incitement to ethnic hatred in December 2024, for comments in a WhatsApp group of journalists, and sentenced her to 21 months in prison.

    CIVICUS Monitor country updates · observed 2025-10-29 · extracted by language model

  • As previously documented, she was convicted in December 2024 by the Mukaza High Court for undermining national integrity and inciting ethnic hatred over comments made in a journalists’ WhatsApp group, and sentenced to 21 months in prison.

    CIVICUS Monitor country updates · observed 2025-10-26 · extracted by language model

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

1 excluded source
  • CIVICUS Monitor country updates, observed 2024-07-01: aged out after 2 years.

Data jurisdiction

What happens to corporate or client data I carry into the jurisdiction?

81Extreme81 of 100, Extreme band, 75 to 100

moderate confidence1 independent source

Why this score (1 source)

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

Surveillance and interception

How comprehensively are communications intercepted, retained, and monitored, and is commercial spyware in use?

69High69 of 100, High band, 50 to 74

moderate confidence2 independent sources

Why this score (3 sources)

  • Freedom House 2026 scored BDI 0 of 4 on freedom to express personal views on sensitive topics without fear of surveillance or retribution.

    Freedom House, Freedom in the World · observed 2026-01-01

  • V-Dem Digital Society Project 2025 scored BDI 2.38 of 4 on how comprehensively the government monitors social media and private online communication, where higher is freer.

    V-Dem Institute, University of Gothenburg, Digital Society Project · observed 2025-12-31

  • V-Dem Digital Society Project 2025 scored BDI 0.54 of 4 on how far privacy protection law shields communications from government access, where higher is freer. This measures what privacy law permits rather than what monitoring is practised, so it is read here as a constraint on interception rather than a measurement of it.

    V-Dem Institute, University of Gothenburg, Digital Society Project · observed 2025-12-31

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

Encryption and VPN legality

Are VPNs and strong encryption lawful, and can I be required to install an application or hand over keys?

68High68 of 100, High band, 50 to 74

moderate confidence1 independent source

No law or enforcement action was found restricting the use of a VPN or strong encryption here. The score below is drawn from general indicators for this country rather than from anything observed about this risk, so it describes an absence of reports rather than evidence that the risk is absent.

What the score rests on (1 indirect source)

  • V-Dem Digital Society Project 2025 scored BDI 1.14 of 4 on whether the government regulates the internet in a way consistent with international human rights standards, where higher is freer. This does not establish whether encryption or VPN use is lawful. It measures internet regulation in general, and is used as indirect evidence because the only survey that reads encryption law directly was published in 2019.

    V-Dem Institute, University of Gothenburg, Digital Society Project · observed 2025-12-31

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

Connectivity continuity

How likely am I to lose access to the internet, or to the specific tools I depend on?

51High51 of 100, High band, 50 to 74

good confidence6 independent sources

Why this score (66 sources)

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

1 excluded source
  • Reporters Without Borders, via a site scoped search, observed 2022-12-02: aged out after 4 years.

Border device risk

How likely is my device to be searched, imaged, or seized at the border, and can I be compelled to unlock it?

49Caution49 of 100, Caution band, 25 to 49

moderate confidence5 independent sources

No reports were found of devices being searched, imaged or seized at this border, or of travellers being compelled to unlock them. The score below is drawn from general indicators for this country rather than from anything observed about this risk, so it describes an absence of reports rather than evidence that the risk is absent.

What the score rests on (5 indirect sources)

Report an inaccuracy (corrections need a link to the reporting, statute or research that shows it).

Network threat

How hostile is the network environment to a foreign visitor, from state actors and criminals alike?

Not assessed

No source currently covers this dimension for this country. That is a gap in our evidence, not a finding that the risk is low.

Trend

02550751002026-08-162026-10-16

The overall score is down 0.2 points over 61 days, from 66.5 to 66.3. Most movement is the event overlay decaying rather than anything happening: a news signal loses half its weight every three weeks, so a score drifts back toward its structural baseline on its own. A step that holds is the one worth reading.

Check this yourself

These are not inputs to the score for Burundi. They are where to look if you want to test what is above against something we did not use.

Sources

This assessment draws on 9 sources: Auswärtiges Amt, Reise- und Sicherheitshinweise; CIVICUS Monitor country updates; Freedom House, Freedom in the World; Georgia Tech Internet Intelligence Lab, Internet Outage Detection and Analysis; Global Affairs Canada, Travel advice and advisories; Global Voices (Advox), via a site scoped search; UK Foreign, Commonwealth and Development Office, Foreign travel advice; US Department of State, Travel Advisories; V-Dem Institute, University of Gothenburg, Digital Society Project. Every contributing claim is quoted with its origin and date above. Model version 0.12.0, computed 2026-10-16.

If something here is wrong, say so. Corrections are handled in public, and we correct because the evidence is wrong, not because someone objects.