Corrections

Travel Risk Calculator publishes information that can affect real travel decisions. If we get something wrong — a coordinate, a country boundary, an outdated advisory link, a baseline score that doesn't reflect current conditions, or an AI-synthesized claim that isn't supported by its cited source — we want to fix it and log the fix.

How to report an error

Three channels, in order of speed:

Feedback widget on the assessment page. If the page shows a "Was this useful?" widget at the bottom, use it. Select "report an issue" and describe the problem. Reports tied to a specific assessment slug are easiest for us to triage.
Contact form. Visit /contact and choose category "Data correction" or "AI claim looks wrong". Include the page URL, the specific sentence or score, and what you believe is correct (with a source if possible).
Email directly. [email protected] works for urgent or sensitive items. Please put "Correction:" at the start of the subject line.

What helps us fix it faster

The exact URL of the page (copy from your browser bar).
The exact wording of the claim or the score value you're disputing.
A source for the correct version — a government advisory, a news article, a recent CDC notice, an authoritative news outlet.
The date and time you saw it (live data feeds refresh on schedules — what you saw may already have been updated when we check).

You don't need to provide all four. Even just "the page says X but it should be Y" is enough to start an investigation.

What we treat as urgent

We aim to investigate within 24 hours and, where possible, fix within 72 hours. We prioritize:

Safety reversals. A claim that says a place is safe when it isn't, or vice versa, that could mislead a traveler.
Outdated emergency numbers, hospital locations, or embassy contacts.
Country/city mismatches. Wrong country attribution (e.g., we once had Dubai listed under Romania — fixed).
AI claims unsupported by their cited snippet.

What we will not change on request

Reduce a threat score because a tourism board, hotel chain, government PR office, or business asks us to.
Remove a city, country, or region from coverage because of political pressure.
Suppress an authentic CDC, GDACS, or government-advisory feed item.
Replace AI-grounded sources with promotional content.

If you believe our scoring methodology itself is wrong (as opposed to a specific data point), see Methodology — that page documents how scores are computed. Methodology changes are tracked in the changelog.

Recent corrections

This log captures notable fixes since launch. It is not exhaustive — small typo and formatting fixes are not logged.

2026-05-14 — Imported 48,060 cities from SimpleMaps World Cities (CC BY 4.0); merged duplicate country entries (e.g., "Korea, South" / "South Korea"); preserved all admin-curated city rows.
2026-05-10 — Fixed Dubai assessments incorrectly attributed to Romania when city was entered without country. Country-resolver now disambiguates by city geography.
2026-05-10 — Bucharest AI briefings no longer fall silent on thin or non-English snippets; added Romanian-language snippet awareness.
2026-05-09 — Compare-cities mode now fetches real-data cards for both cities side-by-side (previously only the destination side rendered).
2026-05-09 — Domestic trips (same-country origin and destination) no longer surface passport/visa/embassy recommendations.
2026-05-09 — AI tier-label discipline fix: a 40/100 score can no longer be described as "high risk" in the synthesized prose. The tier is now computed mechanically and the AI writes around it.

If we can't reach a resolution

We try to respond to every legitimate correction request. If you've reported something and we haven't responded within 5 business days, please email [email protected] with "FOLLOWUP:" at the start of the subject line.