What changed in AI security, Apr 27 to May 3, 2026
Apr 27 to May 3, 2026 (ISO week 2026-W18). Weeks run Monday to Sunday in UTC.
180 records published, +11 on the previous week: 31 vulnerabilities (-21), 0 incidents (no change), 18 research items (+12), 131 news items (+20), 0 policy items (no change).
Critical and high advisories
Vulnerability records rated critical or high, newest first.- High
CVE-2026-7644: A vulnerability has been found in ChatGPTNextWeb NextChat up to 2.16.1. Affected is the function addMcpServer of the…
CVE-2026-7644NVD/CVE Database - High
CVE-2026-6543: IBM Langflow Desktop 1.0.0 through 1.8.4 Langflow allows an attacker to execute arbitrary commands with the privileges…
CVE-2026-6543NVD/CVE Database - High
CVE-2026-4503: IBM Langflow Desktop 1.0.0 through 1.8.4 Langflow could allow an unauthenticated user to view other users' images due…
CVE-2026-4503NVD/CVE Database - High
GHSA-56c3-vfp2-5qqj: n8n-mcp's IPv4-mapped IPv6 addresses bypass SSRF protection in validateUrlSync(), enabling full SSRF for SDK embedders
CVE-2026-42449GitHub Advisory Database - High
GHSA-rch3-82jr-f9w9: Jupyter Notebook Vulnerable to Authentication Token Theft via CommandLinker XSS
CVE-2026-40171GitHub Advisory Database - Critical
GHSA-q5f4-99jv-pgg5: n8n has Prototype Pollution in XML Webhook Body Parser that Leads to RCE
CVE-2026-42231GitHub Advisory Database - High
GHSA-537j-gqpc-p7fq: n8n Vulnerable to XSS via MCP OAuth client
CVE-2026-42235GitHub Advisory Database - High
GHSA-r4v6-9fqc-w5jr: n8n's Credential Authorization Bypass in dynamic-node-parameters Allows Foreign API Key Replay
CVE-2026-42226GitHub Advisory Database - High
GHSA-49m9-pgww-9vq6: n8n Vulnerable to Unauthenticated Denial of Service via MCP Client Registration
CVE-2026-42236GitHub Advisory Database - High
CVE-2026-42249: Ollama for Windows contains a Remote Code Execution vulnerability in its update mechanism due to improper handling of…
CVE-2026-42249NVD/CVE Database - High
CVE-2026-42248: Ollama for Windows does not perform integrity or authenticity verification of downloaded update executables. Unlike…
CVE-2026-42248NVD/CVE Database - High
CVE-2026-7178: A weakness has been identified in ChatGPTNextWeb NextChat up to 2.16.1. This affects the function storeUrl of the file…
CVE-2026-7178NVD/CVE Database - High
CVE-2026-7177: A security flaw has been discovered in ChatGPTNextWeb NextChat up to 2.16.1. Affected by this issue is the function…
CVE-2026-7177NVD/CVE Database - High
CVE-2026-7191- Arbitrary Code Execution via Sandbox Bypass in QnABot on AWS
AWS Security Bulletins
Exploitation signals
Vulnerabilities published in the week that are listed in the CISA Known Exploited Vulnerabilities catalog or have an EPSS score of 10% or more.No vulnerability published in this week is listed as exploited or has an EPSS score of 10% or more.
Packages that began delegating to a language model
Exposure Registry packages whose first release declaring an LLM SDK, agent framework or MCP dependency was published in the week.| Package | Ecosystem | LLM SDKs | Release | Released |
|---|---|---|---|---|
| dottxt | PyPI | OpenAI SDK | 0.2.0 | |
| n8n | npm | LangChain, Model Context Protocol SDK, Pinecone, Qdrant, Vercel AI SDK | 2.19.1 | |
| giskard-llm | PyPI | Anthropic SDK, Google Gemini SDK, OpenAI SDK | 1.0.0b2 |
Topics that moved
Largest increases over the mean of the 4 previous weeks, for topics with at least 3 records in the week.| Topic | Records | Weekly mean, previous 4 | Difference |
|---|---|---|---|
| Inference infrastructure | 3 | 2.8 | +0.3 |
Research
Peer-reviewed first, then newest. Showing 8 of 18.Generating Privacy-Preserving Faces for Multi-Party Secure Authentication
Peer-reviewedIEEE Xplore (Security & AI Journals)Explainability-Guided Untargeted Attacks on Knowledge Graph Embedding
Peer-reviewedIEEE Xplore (Security & AI Journals)Toward Polymorphic Backdoor Against Semantic Communication via Intensity-Based Poisoning
Peer-reviewedIEEE Xplore (Security & AI Journals)Security Enhancement for Person Re-Identification Through Diffusion Driven Semantic Attacks
Peer-reviewedIEEE Xplore (Security & AI Journals)PixOOD: Pixel-Level Out-of-Distribution Detection
Peer-reviewedIEEE Xplore (Security & AI Journals)K-TCDP: A Temporal Correlated DP Mechanism for LoRA Supervised Fine-Tuning
Peer-reviewedIEEE Xplore (Security & AI Journals)From Local to Global to Mechanistic: An iERF-Centered Unified Framework for Interpreting Vision Models
Peer-reviewedIEEE Xplore (Security & AI Journals)Adversary-Robust Graph-Based Learning of WSIs
Peer-reviewedIEEE Xplore (Security & AI Journals)
Policy and regulation
Newest first.No regulatory or policy records were published in this week.
Generated from the AI Sec Watch database at . Every item links to its record.