Privacy UnpackedAnswers from the records
Ask the tracker
Ask about the data breaches on the tracker in plain English. The answer comes only from its records, and every breach it names links to its page.
How it answers
An AI model (Anthropic’s Claude) turns your question into a search of the tracker’s records: by name, date, state or country, what was exposed, how many people, or which ransomware gang. It then writes a short answer from what the search found, naming the records it rests on. Before you see it, the answer is checked by code: every number in it must be in the records the search returned. An answer that fails the check is not shown; you get the matching records instead.
It knows only what the tracker knows: about 5,000 breaches since 2023, mostly from US state attorney-general registers, plus Have I Been Pwned, the news and ransomware gangs’ claims. There is no industry field, so “hospitals” means names with words like hospital or medical in them, and a count is what is on the tracker, not every breach there was. Wondering whether you were affected? Use My breaches. Got a letter? Check it.
What happens to your question
Your question is sent to Anthropic’s API to be answered and is not stored by us: we keep only a count of questions per day. To share the cost fairly, each visitor can ask 12 questions a day; for that, a scrambled form of your internet address is kept in memory for the day and never written down. This page loads no analytics and no ads. Do not type personal details: the tracker has none to look up.
From the data breach tracker: official notices, breach databases and the news, every fact with its source. Open data (CC BY 4.0) · How the tracker works · Official registers · Corrections and changes · Found a mistake? Tell us and we correct it.