Skip to content

Response, New York. Noah Park: Check Point Patches Actively Exploited Management Server Path Traversal. Detection, Dallas. Ava Okello: Report Finds Nearly Half of Deployed Detections Need Attention Before SOCs Can Trust Them.

SOCtember

Always First. Fast SOC News.

Bar chart titled MITRE CVEs Published by Hour of the Week, showing tall spikes at midnight UTC on Monday through Friday, with the label 87 percent of MITRE's 4,509 CVEs are stamped exactly 00:00:00 UTC.
Image: Jerry Gamblin, RogoLabs (fair use). MITRE CVEs by hour on the record over the last 52 weeks, as published by Socket on Nov. 18, 2026.

Tools

MITRE Concedes 87% of Its Recent CVE Records Carry a Placeholder Midnight Timestamp, Skewing Any Tool That Measures How Fast Vulnerabilities Go Public

A researcher found that 3,901 of the 4,509 CVEs MITRE published in the past year were stamped at exactly 00:00:00 UTC, while none of 100,235 records from 425 other numbering authorities were. MITRE blames a legacy client and says it will be phased out in the first quarter of 2027.

James Whitford, ToolsAustin4 min read

AUSTIN - The publication time on most vulnerability records issued by MITRE, the largest single source of CVE identifiers, does not reflect when the records were actually published, the organization has acknowledged, a flaw that quietly distorts dashboards, vendor metrics and research that measure how quickly vulnerability information reaches defenders.

The problem was documented by Jerry Gamblin, founder of the independent vulnerability-intelligence lab RogoLabs, and reported on Wednesday by the software supply chain security company Socket. Looking at CVE records published in the 52 weeks ending Nov. 16, Gamblin found that 3,901 of MITRE's 4,509 records, or 87%, had a datePublished value of exactly 00:00:00.000Z. Across the other 425 CVE Numbering Authorities, none of 100,235 records carried that timestamp.

The CVE Record Format defines datePublished as the date and time a record first appeared in the CVE List, and many tools treat it that way. Gamblin first raised the issue in a GitHub thread in July. "Timestamps are one of the few fields in a CVE Record that are supposed to be objective and machine-trustworthy," he wrote there.

Screenshot of GitHub issue 41 in the CVEProject cve-documents repository, opened by jgamblin on July 7 and titled Require CNA publication via CVE Services API, with accurate timestamps, describing zeroed timestamps on MITRE records and proposing rule changes.
Image: GitHub, via Socket (fair use).Gamblin's July issue asking for accurate CVE publication timestamps.

A 15.8-hour gap that is not a delay

To show the midnight values were placeholders, Gamblin compared each record's stamp with its first appearance in the CVE List's deltaLog.json change log for records published from Oct. 19 to Nov. 16. The 144 MITRE records stamped at midnight surfaced a median of 15.8 hours after their recorded time. The 58 MITRE records with other timestamps appeared a median of 4.6 minutes later, close to the 4.5 minutes for other CNAs. The dates themselves were accurate in his sample; the times were not.

That distinction matters for anyone building on the data. A naive comparison would suggest MITRE takes far longer than other authorities to publish, when the gap is an artifact. Gamblin wrote that anything downstream that measures "time-to-publish, time-to-update, or disclosure latency (EPSS, CNA scorecards, research like mine) is quietly built on bad data for the biggest chunk of the dataset."

Screenshot of a GitHub comment by jgamblin with November data, a table showing 144 MITRE records stamped 00:00:00.000Z with a median of 15.8 hours to first appearance, 58 MITRE records stamped at other times at 4.6 minutes, and 15,546 records from every other CNA at 4.5 minutes.
Image: GitHub, via Socket (fair use).Gamblin's November comparison of recorded timestamps with first appearance in the CVE change log.

MITRE's explanation

Responding on LinkedIn, Alec Summers, MITRE's CVE and CWE project leader, called it "a known historical timestamp issue." He said MITRE publishes through the same CVE Services API as other CNAs, but one endpoint used by some workflows requires the client to supply a time, and "because of concurrency issues in a legacy client architecture, a single predictable value (00:00:00) is best for system stability." He said, "Publications using that legacy architecture will be phased out in Q1 2027."

The shift is already visible in Gamblin's numbers: 97% of MITRE records from 2023 through 2025 carried the midnight stamp, compared with 66% so far in the fourth quarter of 2026. Summers said correcting historical timestamps "has not been highlighted as a demonstrated priority need by CVE data consumers." Gamblin, who sits on CVE Program working groups, welcomed the timeline and asked MITRE to post the plan in his GitHub issue.

What it means for vulnerability tooling

For SOC and vulnerability management teams, the practical step is to stop treating datePublished as precise for MITRE-assigned records. Pipelines that compute exposure windows, compare publication against first exploitation, or score CNAs and vendors on speed should flag midnight-UTC values from MITRE as date-only, or substitute the first appearance in the CVE List change log where timing matters. With CVE volumes rising, Gamblin counted 35,364 CVEs in the first half of 2026, up 49.5% from a year earlier, and VulnCheck has reported that 23.43% of known exploited vulnerabilities in that period showed exploitation on or before their CVE publication date. In that environment, hours of error in a timestamp are not a rounding issue.

About the Author

James Whitford

James Whitford covers SOC tooling, SIEM, SOAR, and detection platforms for SOCtember from Austin.

Related stories

Tools desk