Corrections
Every change made to a record's classification after it was published, with the reason. Reports come from the form on each record page; the classifier audit and the maintainer's own review find the rest. Dataset releases are frozen, so a correction applies to the live site and to the next release.
- Corrections
- 633
- Reports received
- 0
- Reports declined
- 0
- Reports open
- 0
Each correction is listed as its own row. Group bulk changes
| Date | Record | Change | Reason | Found by |
|---|---|---|---|---|
| 2026-10-10 | New ATHR vishing platform uses AI voice agents for automated attacks | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Hackers exploit Marimo flaw to deploy NKAbuse malware from Hugging Face | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | RCE by design: MCP architectural choice haunts AI agent ecosystem | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Cursor AI Vulnerability Exposed Developer Devices | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Copilot & Agentforce offen für Prompt-Injection-Tricks | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Fracturing Software Security With Frontier AI Models | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Anthropic MCP Design Vulnerability Enables RCE, Threatening AI Supply Chain | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Google Patches Antigravity IDE Flaw Enabling Prompt Injection Code Execution | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Prompt injection turned Google’s Antigravity file search into RCE | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Azure SRE Agent flaw lets outsiders silently eavesdrop on enterprise cloud operations | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Google Fixes Critical RCE Flaw in AI-Based Antigravity Tool | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Cohere AI Terrarium Sandbox Flaw Enables Root Code Execution, Container Escape | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Anthropic investigates report of rogue access to hack-enabling Mythos AI | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Anthropic’s most dangerous AI model just fell into the wrong hands | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Toxic Combinations: When Cross-App Permissions Stack into Risk | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Project Glasswing Proved AI Can Find the Bugs. Who's Going to Fix Them? | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Anthropic’s Mythos breach was humiliating | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Bitwarden CLI password manager trojanized in supply chain attack | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | LMDeploy CVE-2026-33626 Flaw Exploited Within 13 Hours of Disclosure | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Discord Sleuths Gained Unauthorized Access to Anthropic’s Mythos | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Microsoft patched an ‘agent-only’ role that was not | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Deepfake Voice Attacks are Outpacing Defenses: What Security Leaders Should Know | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Microsoft Patches Entra ID Role Flaw That Enabled Service Principal Takeover | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Critical Unpatched Flaw Leaves Hugging Face LeRobot Open to Unauthenticated RCE | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Critical Cursor bug could turn routine Git into RCE | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Hackers are exploiting a critical LiteLLM pre-auth SQLi flaw | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Learning from the Vercel breach: Shadow AI & OAuth sprawl | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | New Wave of DPRK Attacks Uses AI-Inserted npm Malware, Fake Firms, and RATs | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Google Fixes CVSS 10 Gemini CLI CI RCE and Cursor Flaws Enable Code Execution | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | SAP npm package attack highlights risks in developer tools and CI/CD pipelines | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Max-severity RCE flaw found in Google Gemini CLI | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Critical Gemini CLI Flaw Enabled Host Code Execution, Supply Chain Attacks | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | The (In)security Landscape of AI-Powered GitHub Actions (Part 2/2) | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Hugging Face, ClawHub Abused for Malware Distribution | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | AI agents can bypass guardrails and put credentials at risk, Okta study finds | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Copirate 365 at DEF CON: Plundering in the Depths of Microsoft Copilot (CVE-2026-24299) | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | We Scanned 1 Million Exposed AI Services. Here's How Bad the Security Actually Is | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Critical Bug Could Expose 300,000 Ollama Deployments to Information Theft | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Supply-chain attacks take aim at your AI coding agents | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Fake Claude AI website delivers new 'Beagle' Windows malware | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Gemini CLI Vulnerability Could Have Led to Code Execution, Supply Chain Attack | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | 'TrustFall' Convention Exposes Claude Code Execution Risk | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Claude Code OAuth Tokens Can Be Stolen Through Stealthy MCP Hijacking | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | When prompts become shells: RCE vulnerabilities in AI agent frameworks | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Ollama vulnerability highlights danger of AI frameworks with unrestricted access | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Vulnerability in Claude Extension for Chrome Exposes AI Agent to Takeover | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Claude in Chrome is taking orders from the wrong extensions | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Fake OpenAI repository on Hugging Face pushes infostealer malware | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Ollama Out-of-Bounds Read Vulnerability Allows Remote Process Memory Leak | Severity: critical to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |
| 2026-10-10 | Hackers abuse Google ads, Claude.ai chats to push Mac malware | Severity: high to medium | Severity capped at medium. High and critical measure technical exploitability and are reserved for vulnerabilities and incidents; this record is a news report. | Classifier audit |