Method
How an incident ends up on the globe
A short, honest account of where HackMap's data comes from, how each story is judged, and where the map's limits are.
HackMap is a map of reported cybersecurity incidents, not a live sensor network. Every point on the globe traces back to a published article, and the source link sits on the card for each incident. Understanding the pipeline below is the best way to read the map accurately.
1. Collection from public reporting
A collector reads RSS feeds from established security desks and general technology newsrooms several times a day. For general technology feeds, only stories that read like an outage, disruption or company-caused failure are kept, because most of their output is product news rather than incidents. Stories older than a few days are skipped, and any article already on the map is ignored so the same event is not plotted twice.
2. Classification into ten incident types
Each candidate story is examined against a fixed set of criteria. It counts as an incident only when it reports a specific event against a named organisation, service or product: a data breach or leak of stolen data, a ransomware or other targeted cyberattack, a crypto or DeFi exploit, an AI-related attack such as model abuse or deepfake fraud, a DDoS campaign that knocked a target offline, espionage or APT intrusion, a fraud and scam operation with a named victim, an actively exploited zero-day in named software, a service outage at a major provider, or a company's own misconfiguration, bad update or accidental exposure.
Opinion pieces, patch advisories with no victim, product launches, funding news, generic security advice, maintenance notices, roundups that name no single victim and law-enforcement arrests without a clear victim are deliberately excluded. Requests to classify something that is not news are ignored as well, which keeps the map free of injected or promotional items.
3. Placing the pin
Each incident is assigned the city and country of the affected organisation's headquarters or most relevant office, along with decimal coordinates, and the summary is written in two or three sentences for readers who already know security basics. When the location genuinely cannot be established, the incident is left off the map rather than tagged with a vague "global" marker, so a missing pin usually means an unknown location rather than a missing story.
4. What the map does not show
Volume on this globe reflects press coverage, not the true frequency of attacks: well reported regions and large brands appear more often than quiet ones. Attribution in summaries reflects what the source reported, not an independent investigation. The map carries a rolling window of recent incidents, so an older event drops off even though it still matters to the organisation involved. Nothing here is a substitute for an advisory from the vendor or a national CERT, and it should not be used to rank one company's security against another's.
5. Verifying an incident yourself
Every incident card links to the article it came from. If you need to act on something you see here, open that link first: check the publication, the date, whether the vendor has confirmed the event, and whether the reporting has since been updated or retracted. Corrections and better sources are always welcome through the contact page.