🔍 Read the full analysis: CVEs Are Going To The Moon: What The Data Shows About AI, Bug Hunters And The Patch Gap on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
The source analysis reports 88,772 CVE records published in 2026 through October 25, 78% more than in all of 2025. It says the rise reflects concentrated disclosure activity and growing bug discovery, but public records do not show how much AI contributed; the author also reports that only 516 of 6,157 vulnerabilities Anthropic said it reported to maintainers were patched to its knowledge.
CVE records have risen sharply in 2026, with 88,772 published through October 25, according to an analysis of US National Vulnerability Database and CVE.org records by Thorsten Meyer. That is 78% more than the 50,000 records reported for all of 2025 in the source analysis, but the available data cannot establish how much artificial intelligence contributed to finding the flaws or how many are being exploited.
Between October 1 and 25, 2026, 13,934 new CVE records were published, the analysis says. That is 3.2 times the 4,389 records published during all of October 2025. The monthly rate has climbed since June, though the report says its growth slowed: daily volume rose 20% from June to July, 28% from July to August and 22% from August to September, then about 11% in October through the 25th.
The increase is concentrated among a small number of organizations. Microsoft had 1,455 October records, compared with 180 in October 2025; Oracle rose from 78 to 1,078, most on October 20. GitHub and VulnCheck also recorded large increases. The report says the Linux kernel count rose less sharply, from 660 to 925. These figures count records by publication date, not necessarily newly discovered bugs.
The analysis says 54% of October records did not name a finder. Two hundred records, or 1.4%, explicitly credited an AI lab, AI model or AI security firm. The author describes that as a minimum because attribution is often absent. Separately, Anthropic’s dashboard, as of October 2, listed 6,157 vulnerabilities reported to maintainers; the company said 516 were patched to its knowledge. That reported-to-patched gap points to a separate challenge from discovery: getting fixes completed and deployed.
Infographic 01 · How big the CVE wave really is
The CVE chart is real. Steep, not parabolic.
New vulnerability records have tripled year over year. A handful of big vendors and clearinghouses drive the jump, and the growth rate cooled in October.
New CVE records per month
January 2024 – 25 October 2026 · phones show the last 13 months
Growth of new CVEs per day versus the previous month. Steep, but the acceleration slowed in October: not “parabolic”.
Who publishes the jump
records by CVE Numbering Authority, October vs. October
Microsoft’s October Patch Tuesday fixed a record 1,413 CVEs (SecurityWeek (14 Oct 2026)). The Linux kernel grew far less.
Not a backlog dump
checks on the October 2026 records
- 98% carry a 2026 ID
- 15 days median from reserving an ID to publishing it
- 217 numbering authorities published, vs. 163 in Oct 2025
- 202 records were later rejected (counted, as NVD does)
- ~17,000 if October’s pace holds: a projection, not a count
Sources: NVD CVE API 2.0 (counts by publication date, all statuses, queried 26 Oct 2026); CVE.org records (credits field, fetched 26 Oct 2026); SecurityWeek (14 Oct 2026). Own analysis; raw files and queries in the kit.
Powered by Thorsten Meyer AI
Discovery Is Outrunning Fixes
A rise in published vulnerabilities can reflect more bugs being found, more disclosures being processed, or both. The increase matters to software makers and users because each record can require triage, a fix, testing and deployment across products. But a CVE count is not a count of successful attacks, nor does it show how many systems remain exposed.
The patch figures cited in the source make the difference between finding and fixing concrete. Anthropic reported thousands of vulnerabilities to maintainers, while saying it knew of patches for 516. That does not establish that the remaining reports are all unresolved: maintainers may have fixed issues without the dashboard reflecting it, or reports may require further review. Still, the figures underscore why patch availability and adoption are central to the security impact of a discovery surge.
AI’s role also matters for how organizations plan defenses, but the available evidence is limited. The analysis finds explicit AI attribution in a small share of October records, while noting that most records omit finder information. Readers should not treat the record totals as proof that AI caused the surge or that AI is driving a corresponding wave of attacks.
Products Worth Considering

Samsung 27-Inch Odyssey G5 QHD Gaming Monitor (G50F)

Apple AirPods Pro (2nd Generation) Wireless Earbuds with USB-C MagSafe Charging Case

LG 27GS60QC-B Ultragear 27-inch Curved Gaming Monitor (QHD, 180Hz, 1ms, 1000R)
As an affiliate, we earn on qualifying purchases.
How the CVE Counts Were Built
Meyer says he queried the NVD CVE API 2.0 for records published from January 2022 onward, counting by publication date and retrieving the data on October 26, 2026. He cross-checked October records against CVE.org. The source says the two records systems agreed on 13,766 October records, plus 202 records later rejected; its separate October 1–25 count is 13,934 new records.
A CVE is a public identifier for a disclosed vulnerability. The source describes publication as part of coordinated disclosure, usually handled by a CVE Numbering Authority such as a vendor. That process records vulnerabilities disclosed to the public; it does not identify every person or tool involved in finding them. In October, 6,346 records named a finder, according to the analysis, leaving more than half without one.
The report also cites Anthropic’s vulnerability dashboard and OpenAI’s Patch the Planet program with Trail of Bits. Anthropic reported that external firms reviewed 6,123 candidate vulnerabilities identified by its Claude models and confirmed 92.7% as real; the company said 6,157 had been reported to maintainers. OpenAI and Trail of Bits reported “hundreds of security issues” across 19 open-source projects, with each reviewed by people. These are company-reported program figures, not a full accounting of AI’s share of all CVEs.
“6,157 vulnerabilities have been reported to maintainers.”
— Anthropic, as reported in its disclosure dashboard
What the Records Cannot Show
The published counts do not establish how many vulnerabilities AI found. The source says 54% of October records list no finder, and only 1.4% explicitly credit an AI lab or security firm. The latter figure is therefore a count of explicit credits, not a reliable estimate of AI’s total contribution.
The source material also does not provide a complete October comparison for confirmed exploitation. It begins to cite CISA’s Known Exploited Vulnerabilities catalog, saying the catalog added 245 entries during 2025, but the supplied text ends before giving the relevant 2026 figures. The article therefore cannot compare the CVE surge with a full, current exploitation measure.
It is also unclear how many of the vulnerabilities reported to maintainers but not listed as patched have since been fixed, remain under review, or are affected by other status changes. The reported numbers are snapshots, and the October data covers only records published through the 25th.
Tracking Patches and Exploitation
The next useful checks are updated CVE totals after October closes, subsequent status information from Anthropic and affected maintainers, and a complete comparison with CISA’s exploited-vulnerability catalog. Those measures could help distinguish a rise in disclosure activity from a rise in vulnerabilities actively abused by attackers.
For now, the source analysis projects that October could finish near 17,000 records if its pace through October 25 continues. That is the author’s projection, not a final count. The main open question is whether organizations can verify, prioritize and deploy fixes as quickly as new vulnerabilities are disclosed.
Key Questions
How many CVE records were published in 2026?
The source analysis counts 88,772 through October 25, 2026, or 78% more than in all of 2025, using its stated year-to-date comparison.
Does the surge prove AI is finding most vulnerabilities?
No. The analysis says only 1.4% of October records explicitly credited an AI lab, model or AI security firm, while 54% named no finder. Those gaps mean the records cannot establish AI’s overall share.
Are more CVEs evidence of more active attacks?
Not by themselves. CVE records document public vulnerability disclosures; they do not show whether a flaw is being exploited. The supplied source material does not include a complete 2026 CISA exploited-vulnerability comparison.
What does the Anthropic patch figure mean?
Anthropic’s dashboard said 6,157 vulnerabilities had been reported to maintainers. Meyer said 516 were patched to the company’s knowledge. The figures do not establish that every other report remains unfixed, since status information may be incomplete or change over time.
Source: ThorstenMeyerAI.com
