The US National Institute of Standards and Technology (NIST) unveiled a new metric called Likely Exploited Vulnerabilities (LEV) to improve the assessment of whether software vulnerabilities have been exploited in the wild. LEV builds upon the Exploit Prediction Scoring System (EPSS), a data-driven model that predicts the likelihood a vulnerability will be exploited within a 30-day window. EPSS, originally launched in 2018, was updated to version 4 in March 2025, incorporating more refined scoring methods.
LEV offers a more nuanced perspective by providing daily data on each CVE, including the probability of past exploitation, peak EPSS scores, and a history of EPSS scores over 30-day periods. This approach enables vulnerability managers to track exploitation trends over time and understand each vulnerability’s history in greater detail. LEV also includes affected product details using Common Platform Enumeration (CPE) values, supporting better decision-making in vulnerability prioritization.
Designed to complement both EPSS and Known Exploited Vulnerability (KEV) lists from sources such as CISA and private sector providers, LEV helps fill gaps left by these tools. NIST highlights that KEV lists are not fully comprehensive and that EPSS scores do not factor in past exploitation events, which can limit their predictive accuracy. LEV addresses these shortcomings by incorporating historical exploitation data to improve prioritization efforts.
Despite its benefits, LEV faces limitations, including an unknown margin of error primarily due to the reliance on EPSS data, which does not adjust scores for vulnerabilities exploited within the previous 30 days. NIST acknowledges these constraints but expects the LEV metric and the accompanying white paper to spur improvements in how exploited vulnerabilities are tracked and managed, ultimately strengthening cybersecurity defenses for organizations.
Reference:
