PulseAugur
EN
LIVE 19:29:56

52 prompt injection CVEs registered in LLM software by November 2026

By PulseAugur Editorial · [1 sources] ·

As of November 6, 2026, the National Vulnerability Database has recorded 52 CVEs related to prompt injection in LLM and agent software. These vulnerabilities, predominantly categorized as high or critical severity with a median CVSS score of 8.0, have seen a significant increase in reported incidents since September. Coding agents, AI gateways, and MCP servers are among the most affected product types, often involving classic vulnerabilities like server-side request forgery or missing authorization as entry points for prompt injection. AI

IMPACT Highlights the growing and quantifiable security risks associated with prompt injection in LLM applications, emphasizing the need for robust defenses.

RANK_REASON The item details a quantitative analysis of registered vulnerabilities (CVEs) related to prompt injection in LLM software, including severity, timeline, product types, and weakness classes, based on data from the National Vulnerability Database. [lever_c_demoted from research: ic=1 ai=1.0]

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

52 prompt injection CVEs registered in LLM software by November 2026

How we ranked this

Signal score
48 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The item details a quantitative analysis of registered vulnerabilities (CVEs) related to prompt injection in LLM software, including severity, timeline, product types, and weakness classes, based o…
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
safety, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Gökhan Köse ·

    52 prompt-injection CVEs in 2026 so far: what the numbers say

    <p>Most articles about prompt injection quote the same two numbers: OWASP ranks it first in the LLM Top 10, and "most models are vulnerable". Neither tells you how often it actually turns into a registered vulnerability. So we counted.</p> <p>Since January 2026 the National Vulne…

RELATED ENTITIES

RELATED TOPICS