Exploitability in practice
Global signals answer whether attackers use a flaw anywhere. KEV requires reliable evidence of exploitation in the wild, and FIRST's EPSS team notes that only around 5% of all vulnerabilities are ever exploited in the wild. Those numbers make exploitation evidence a strong filter for a backlog measured in tens of thousands of findings.
Local exploitability is what validation measures: can the flaw be reached from an entry point, does a control block the technique, does the asset run the vulnerable configuration. Gartner's CTEM guidance puts this first in the validation stage: confirm attackers could actually exploit a vulnerability. A KEV-listed flaw behind a working control may rank below an unlisted one on an exposed path.
An example
CISA's KEV JSON feed, catalog version 2026.11.10 released November 10, 2026, lists 1,779 vulnerabilities with evidence of exploitation in the wild.
Source, read 12 Nov 2026.
Related terms
Sources
- https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json, read 12 Nov 2026
- https://www.first.org/epss/articles/prob_percentile_bins, read 12 Nov 2026
- https://web.archive.org/web/20260110234446/https://www.gartner.com/en/articles/how-to-manage-cybersecurity-threats-not-episodes, read 12 Nov 2026
Last reviewed 12 Nov 2026.