Research
Six questions the corpus can answer, and what each answer does and does not support.
Every chart here is built from our own pipeline — NVD and EPSS ingest, CISA KEV, per-CVE annotations and our cross-walks, each mapping labelled by how it was checked. Organised by the question you arrived with rather than by the table the data came from.
Last updated: 2026-11-20 16:03 UTC
Volume is rising and severity is not falling with it, so the corpus is getting bigger without getting milder — but publication volume measures disclosure, not risk, and a quiet month is not a safe one.
CVE publication volume — monthly
CVSS distribution by quarter
Substantively-modified CVEs — weekly
Where this comes from. Built from the CVE corpus this pipeline ingests (NVD, EPSS, and our per-CVE annotations). These are publication-side signals: what was disclosed, when, and at what severity. A rising line here is not evidence of rising attacks — that question is the next one.
KEV is the only confirmed-exploited signal here and it is a floor, not a census: it counts what CISA has confirmed, so everything else on this tab is an estimate of pressure rather than a count of attacks.
CISA KEV — additions per month
Time from publication to KEV listing
Ransomware-attributed share of new KEV entries
Exploit pressure of the CVE corpus
Attacker-attributed CVEs — weekly
cve_actor_attribution. Coverage is partial — the
annotator runs daily over CVEs with synthesised summaries; recent
weeks may grow as the backfill catches up.
Where this comes from. Built from CISA's KEV catalog (confirmed exploited in the wild), EPSS probabilities, and ransomware flags on KEV entries. This is not attacker telemetry. It measures what is publicly known to be exploited, so it lags real exploitation and undercounts anything nobody reported.
The weakness mix moves slowly, which is the finding — the same handful of CWEs stay at the top year after year, so a defence tuned to last year's ranking is mostly still tuned to this year's.
Top weaknesses (CWE) — rank shift across years
ATT&CK techniques mapped to recent CVEs — rank shift
MITRE ATT&CK v20.0
Tactic coverage of recent CVEs (Σ EPSS)
Where this comes from. Built from CWE assignments on published CVEs and our CWE→ATT&CK cross-walks. Rank shifts are shifts in what gets assigned; a weakness class can move because analyst practice changed rather than because attackers did.
Vendor CVE counts measure disclosure practice as much as product quality: a vendor that publishes diligently outranks one that stays quiet, so read high counts as visibility, not as a verdict.
Vendor cohorts — monthly CVE volumeAI
Reserved — Customer asset coverage
The auth-service’s “My Environment” feature already maps user-uploaded CPEs to active CVEs. When we surface aggregate (anonymised) statistics — which vendors appear most in user portfolios, average CVE exposure per portfolio — this slot becomes a real chart.
Per-vendor sparklines — monthly CVE counts
Top-15 vendors by KEV adds (quarterly)
Product categories compared — six metrics side by side
Where this comes from. Built from CPE product matches on CVEs, with CNA-assignment fallback where CPE data lags. Vendors ship different amounts of software and disclose at different rates, so volume here reads as attention and surface, not as relative product quality.
One chart, honestly. This is the thinnest evidence on the page: it shows which control families our mappings attach to recent CVEs, which is a statement about our cross-walks before it is a statement about your controls.
Control family share of mitigations
Reserved — Configuration rule coverage by control family
Coming when CIS Benchmarks / AWS Config conformance / Azure Policy / STIG ingestion lands. See the controls catalogue for the 49 cloud-native rules we have today.
Reserved — Implementation drift
Coming when configuration-drift observations are ingested. Will chart deviation between a customer's running configuration and their declared baseline, by control family.
Where this comes from. Built from our control↔weakness cross-walks, which are authored and graded rather than observed. It shows which control families carry the mitigation weight on paper. Whether a control is deployed, tuned, or working is not something this data can see.
This is the one argument on the page rather than a measurement. The difference-in-differences compares an LLM-credited cohort against a matched one; it supports a claim about severity mix in the credited cohort, and it does not establish that LLMs caused the difference.
Severity impact of LLM-assisted vulnerability discoveryAI
Section A — Direct attribution (high-confidence, narrow)
3056 CVEs detected with explicit LLM-discovery credit.
By LLM family: Openai: 2347, Anthropic: 863, Moonshot: 3, Google: 2, Alibaba: 1, Deepseek: 1.
Specific models cited (top 8): OpenAI (2342), Anthropic (539), Claude (226), Project Glasswing (23), Claude Code (6), Kimi (3), GPT5 (2), Claude Opus 4.6 (2).
| Metric | Mythos-credited (n / mean) | All other CVEs (n / mean) | Difference | 95% CI | p (Holm) | Hedges’ g / RD |
|---|---|---|---|---|---|---|
| CVSS base | 3,056 / 7.24 | 120,126 / 6.85 | +0.394 | [+0.326, +0.463] | 1.9e-43 | +0.231 |
| EPSS | 3,056 / 0.0034 | 120,126 / 0.0071 | -0.004 | [-0.004, -0.003] | 2.8e-29 | -0.093 |
| KEV-listing rate | 3,056 / 0.0% | 120,126 / 0.3% | -0.003 | [-0.003, -0.002] | 0.003 | -0.003 |
| Risk Priority | 3,056 / 53.26 | 120,126 / 51.17 | +2.090 | [+1.653, +2.526] | 1e-33 | +0.180 |
Section B — Cohort × time-window DiD
| Metric | Treated pre (n / mean) | Treated post (n / mean) | Control pre (n / mean) | Control post (n / mean) | DiD | 95% CI | p (Holm) | Hedges’ g / RD |
|---|---|---|---|---|---|---|---|---|
| CVSS base | 8,080 / 6.67 | 15,160 / 7.13 | 40,708 / 6.75 | 59,234 / 6.89 | +0.331 | [+0.285, +0.377] | 7.3e-109 | +0.296 |
| EPSS | 8,080 / 0.0087 | 15,160 / 0.0054 | 40,708 / 0.0100 | 59,234 / 0.0051 | +0.002 | [+0.000, +0.003] | 1.1e-58 | -0.078 |
| KEV-listing rate | 8,080 / 0.9% | 15,160 / 0.4% | 40,708 / 0.3% | 59,234 / 0.2% | -0.005 | [-0.007, -0.002] | 2.3e-07 | -0.006 |
| Risk Priority | 8,080 / 49.56 | 15,160 / 52.41 | 40,708 / 50.94 | 59,234 / 51.33 | +2.461 | [+2.141, +2.780] | 2.7e-94 | +0.265 |
Distribution split — cohort × period
Per-vendor: more vulns AND more severe?
Robustness — matched 3-week pre/post window
| Metric | Treated pre (n / mean) | Treated post (n / mean) | Control pre (n / mean) | Control post (n / mean) | DiD | 95% CI | p (Holm) | Hedges’ g / RD |
|---|---|---|---|---|---|---|---|---|
| CVSS base | 623 / 6.57 | 799 / 7.04 | 3,882 / 6.95 | 2,802 / 6.67 | +0.740 | [+0.557, +0.920] | 8.4e-09 | +0.306 |
| EPSS | 623 / 0.0031 | 799 / 0.0073 | 3,882 / 0.0085 | 2,802 / 0.0064 | +0.006 | [+0.002, +0.011] | 0.0003 | +0.103 |
| KEV-listing rate | 623 / 0.3% | 799 / 0.6% | 3,882 / 0.2% | 2,802 / 0.1% | +0.004 | [-0.003, +0.012] | 0.48 | +0.003 |
| Risk Priority | 623 / 48.59 | 799 / 51.15 | 3,882 / 52.08 | 2,802 / 50.08 | +4.570 | [+3.326, +5.794] | 0.0003 | +0.245 |
Parallel-trends check (pre-paper)
Where this comes from. Built from acknowledgement text on published CVEs, which is where AI-assisted discovery is credited when it is credited at all. Attribution is voluntary and inconsistent, so the level is a floor rather than a count — which is why the charts here compare cohorts over time instead of trusting any single total.
Active anomalies
ATT&CK T1587.004 rising 79% YoY: 3,466 CVE-associations in H1 2026 vs 1,933 in H1 2025
ATT&CK T1559.003 rising 108% YoY: 3,531 CVE-associations in H1 2026 vs 1,699 in H1 2025
ATT&CK T1584 rising 73% YoY: 6,037 CVE-associations in H1 2026 vs 3,498 in H1 2025
ATT&CK T1599 rising 62% YoY: 5,300 CVE-associations in H1 2026 vs 3,272 in H1 2025
ATT&CK T1584.004 rising 59% YoY: 7,251 CVE-associations in H1 2026 vs 4,556 in H1 2025
Σ EPSS of newly-published CVEs rising 102% YoY: 68.4 this month (17,090 CVEs) vs 33.9 same month prior year (4,273 CVEs)
ATT&CK T1211 rising 51% YoY: 7,106 CVE-associations in H1 2026 vs 4,711 in H1 2025
ATT&CK T1499 rising 93% YoY: 11,945 CVE-associations in H1 2026 vs 6,184 in H1 2025
CWE-125 rising 77% YoY: 946 CVEs in H1 2026 vs 534 in H1 2025
CWE-119 rising 30% YoY: 758 CVEs in H1 2026 vs 585 in H1 2025
CWE-601 rising 61% YoY: 203 CVEs in H1 2026 vs 126 in H1 2025
CWE-190 rising 102% YoY: 354 CVEs in H1 2026 vs 175 in H1 2025
CWE-284 rising 96% YoY: 879 CVEs in H1 2026 vs 449 in H1 2025
CWE-94 rising 76% YoY: 573 CVEs in H1 2026 vs 326 in H1 2025
CWE-200 rising 71% YoY: 651 CVEs in H1 2026 vs 381 in H1 2025
CWE-416 rising 139% YoY: 1,408 CVEs in H1 2026 vs 590 in H1 2025
ATT&CK T1212 rising 78% YoY: 13,087 CVE-associations in H1 2026 vs 7,350 in H1 2025
CWE-770 rising 114% YoY: 469 CVEs in H1 2026 vs 219 in H1 2025
CWE-77 rising 65% YoY: 569 CVEs in H1 2026 vs 344 in H1 2025
CWE-287 rising 110% YoY: 372 CVEs in H1 2026 vs 177 in H1 2025
CWE-400 rising 127% YoY: 486 CVEs in H1 2026 vs 214 in H1 2025
CWE-269 rising 118% YoY: 329 CVEs in H1 2026 vs 151 in H1 2025
ATT&CK T1190 rising 51% YoY: 21,968 CVE-associations in H1 2026 vs 14,528 in H1 2025
ATT&CK T1210 rising 75% YoY: 14,969 CVE-associations in H1 2026 vs 8,563 in H1 2025
ATT&CK T1499.004 rising 88% YoY: 11,657 CVE-associations in H1 2026 vs 6,185 in H1 2025
ATT&CK T1068 rising 63% YoY: 12,901 CVE-associations in H1 2026 vs 7,906 in H1 2025
CWE-306 rising 193% YoY: 466 CVEs in H1 2026 vs 159 in H1 2025
CWE-401 falling 19% YoY: 329 CVEs in H1 2026 vs 408 in H1 2025
CWE-285 rising 185% YoY: 211 CVEs in H1 2026 vs 74 in H1 2025
CWE-266 rising 47% YoY: 313 CVEs in H1 2026 vs 213 in H1 2025
CWE-362 rising 106% YoY: 447 CVEs in H1 2026 vs 217 in H1 2025
CWE-434 rising 7% YoY: 347 CVEs in H1 2026 vs 325 in H1 2025
CWE-120 rising 0% YoY: 410 CVEs in H1 2026 vs 409 in H1 2025
CWE-122 rising 131% YoY: 482 CVEs in H1 2026 vs 209 in H1 2025
CWE-502 rising 73% YoY: 555 CVEs in H1 2026 vs 320 in H1 2025
CWE-295 rising 166% YoY: 205 CVEs in H1 2026 vs 77 in H1 2025
CWE-476 falling 23% YoY: 663 CVEs in H1 2026 vs 860 in H1 2025
CWE-121 rising 158% YoY: 461 CVEs in H1 2026 vs 179 in H1 2025
CWE-639 rising 247% YoY: 579 CVEs in H1 2026 vs 167 in H1 2025
CWE-863 rising 106% YoY: 595 CVEs in H1 2026 vs 289 in H1 2025
CWE-352 falling 54% YoY: 597 CVEs in H1 2026 vs 1,304 in H1 2025
CWE-98 rising 102% YoY: 507 CVEs in H1 2026 vs 251 in H1 2025
CWE-20 rising 229% YoY: 1,001 CVEs in H1 2026 vs 304 in H1 2025
CWE-918 rising 187% YoY: 784 CVEs in H1 2026 vs 273 in H1 2025
CWE-78 rising 156% YoY: 809 CVEs in H1 2026 vs 316 in H1 2025
CWE-22 rising 143% YoY: 1,362 CVEs in H1 2026 vs 561 in H1 2025
CWE-862 rising 54% YoY: 1,737 CVEs in H1 2026 vs 1,130 in H1 2025
Past anomalies (resolved)
Anomalies that triggered on a previous run but no longer do. Showing the most recent 50.
ATT&CK T1548 rising 94% YoY: 3,079 CVE-associations in H1 2026 vs 1,586 in H1 2025
ATT&CK T1185 rising 96% YoY: 3,305 CVE-associations in H1 2026 vs 1,690 in H1 2025
Σ EPSS of newly-published CVEs rising 102% YoY: 68.8 this month (14,954 CVEs) vs 34.1 same month prior year (4,321 CVEs)
KEV additions accelerating 169% YoY: 43 CVEs added to CISA KEV in 2026-09 vs 16 in 2025-09
ATT&CK T1566 rising 60% YoY: 18,798 CVE-associations in H1 2026 vs 11,721 in H1 2025
ATT&CK T1574 rising 60% YoY: 20,306 CVE-associations in H1 2026 vs 12,697 in H1 2025
ATT&CK T1505 rising 57% YoY: 17,422 CVE-associations in H1 2026 vs 11,077 in H1 2025
ATT&CK T1495 rising 50% YoY: 17,708 CVE-associations in H1 2026 vs 11,798 in H1 2025
ATT&CK T1564.014 rising 51% YoY: 14,638 CVE-associations in H1 2026 vs 9,675 in H1 2025
ATT&CK T1007 rising 54% YoY: 14,782 CVE-associations in H1 2026 vs 9,593 in H1 2025
ATT&CK T1203 rising 68% YoY: 14,847 CVE-associations in H1 2026 vs 8,854 in H1 2025
ATT&CK T1134.004 rising 59% YoY: 13,169 CVE-associations in H1 2026 vs 8,275 in H1 2025
ATT&CK T1027.008 rising 65% YoY: 11,637 CVE-associations in H1 2026 vs 7,056 in H1 2025
ATT&CK T1134 rising 50% YoY: 11,936 CVE-associations in H1 2026 vs 7,950 in H1 2025
ATT&CK T1059.010 rising 62% YoY: 10,297 CVE-associations in H1 2026 vs 6,370 in H1 2025
ATT&CK T1082 rising 112% YoY: 3,907 CVE-associations in H1 2026 vs 1,846 in H1 2025
ATT&CK T1498.001 rising 141% YoY: 509 CVE-associations in H1 2026 vs 211 in H1 2025
ATT&CK T1498.002 rising 143% YoY: 513 CVE-associations in H1 2026 vs 211 in H1 2025
ATT&CK T1498 rising 143% YoY: 513 CVE-associations in H1 2026 vs 211 in H1 2025
ATT&CK T1499.001 rising 139% YoY: 514 CVE-associations in H1 2026 vs 215 in H1 2025
ATT&CK T1499.002 rising 151% YoY: 544 CVE-associations in H1 2026 vs 217 in H1 2025
ATT&CK T1034 rising 155% YoY: 789 CVE-associations in H1 2026 vs 310 in H1 2025
ATT&CK T1059.003 rising 99% YoY: 2,016 CVE-associations in H1 2026 vs 1,015 in H1 2025
ATT&CK T1059.001 rising 100% YoY: 2,040 CVE-associations in H1 2026 vs 1,019 in H1 2025
ATT&CK T1687 rising 66% YoY: 1,092 CVE-associations in H1 2026 vs 656 in H1 2025
ATT&CK T1499.003 rising 77% YoY: 1,123 CVE-associations in H1 2026 vs 634 in H1 2025
ATT&CK T1133 rising 199% YoY: 789 CVE-associations in H1 2026 vs 264 in H1 2025
ATT&CK T1552.001 rising 56% YoY: 610 CVE-associations in H1 2026 vs 392 in H1 2025
ATT&CK T1557 rising 145% YoY: 673 CVE-associations in H1 2026 vs 275 in H1 2025
ATT&CK T1078 rising 68% YoY: 778 CVE-associations in H1 2026 vs 462 in H1 2025
ATT&CK T1059 rising 81% YoY: 2,562 CVE-associations in H1 2026 vs 1,414 in H1 2025
ATT&CK T1059.004 rising 100% YoY: 2,042 CVE-associations in H1 2026 vs 1,021 in H1 2025
ATT&CK T1027 rising 207% YoY: 946 CVE-associations in H1 2026 vs 308 in H1 2025
ATT&CK T1036.001 rising 191% YoY: 1,173 CVE-associations in H1 2026 vs 403 in H1 2025
ATT&CK T1543 rising 58% YoY: 1,155 CVE-associations in H1 2026 vs 731 in H1 2025
ATT&CK T1021.002 rising 85% YoY: 1,232 CVE-associations in H1 2026 vs 667 in H1 2025
ATT&CK T1021 rising 85% YoY: 1,265 CVE-associations in H1 2026 vs 684 in H1 2025
ATT&CK T1553.004 rising 108% YoY: 1,266 CVE-associations in H1 2026 vs 610 in H1 2025
ATT&CK T1556.006 rising 108% YoY: 1,266 CVE-associations in H1 2026 vs 610 in H1 2025
ATT&CK T1505.005 rising 82% YoY: 1,268 CVE-associations in H1 2026 vs 698 in H1 2025
ATT&CK T1134.001 rising 93% YoY: 1,274 CVE-associations in H1 2026 vs 659 in H1 2025
ATT&CK T1550.004 rising 102% YoY: 1,317 CVE-associations in H1 2026 vs 653 in H1 2025
ATT&CK T1552.004 rising 67% YoY: 1,346 CVE-associations in H1 2026 vs 805 in H1 2025
ATT&CK T1548.001 rising 99% YoY: 1,775 CVE-associations in H1 2026 vs 894 in H1 2025
ATT&CK T1548.006 rising 109% YoY: 2,041 CVE-associations in H1 2026 vs 977 in H1 2025
ATT&CK T1548.003 rising 109% YoY: 2,385 CVE-associations in H1 2026 vs 1,139 in H1 2025
ATT&CK T1556 rising 91% YoY: 2,871 CVE-associations in H1 2026 vs 1,500 in H1 2025
ATT&CK T1083 rising 98% YoY: 1,630 CVE-associations in H1 2026 vs 825 in H1 2025
ATT&CK T1005 rising 54% YoY: 1,231 CVE-associations in H1 2026 vs 798 in H1 2025
Σ EPSS of newly-published CVEs rising 49% YoY: 59.3 this month (12,308 CVEs) vs 39.8 same month prior year (3,629 CVEs)