Mindgard
PulseAugur coverage of Mindgard — every cluster mentioning Mindgard across labs, papers, and developer communities, ranked by signal.
- 2026-08-12 funding Mindgard secured $30 million to scale its AI security platform. source
4 day(s) with sentiment data
-
AI Security Firm Mindgard Raises $30M; Raycast Enhances AI Chat with Screen Awareness
Mindgard has secured $30 million in funding to expand its AI security platform, which emphasizes continuous testing and simulated real-world attacks. Concurrently, Raycast has launched a new feature called Screen Awaren…
-
Cursor IDE vulnerability allows execution of disguised git.exe
A proof-of-concept (PoC) demonstrated a vulnerability in Cursor version 3.2.16 on Windows, where opening a malicious repository could cause the application to execute a disguised git.exe file located in the repository's…
-
AI vulnerability disclosure model is inconsistent and risky
The current model for disclosing vulnerabilities in AI systems is inconsistent and poses risks as businesses increasingly adopt AI. Unlike traditional software, AI vulnerability findings may not fit neatly into establis…
-
Cursor IDE's critical zero-day vulnerability remains unpatched after 7 months
A critical zero-day vulnerability in the AI-powered code editor Cursor allows for arbitrary code execution when a user opens a malicious project. The vulnerability, discovered by Mindgard, involves a planted `git.exe` i…
-
AI startup Mindgard finds flaw in ChatGPT's safety filters
A British AI security startup, Mindgard, has found a way to bypass ChatGPT's safety filters. By instructing the model to describe an image that is not provided, users can trick ChatGPT into generating inappropriate content.
-
ChatGPT image filters bypassed, generating violent and explicit content
Research from Mindgard has revealed a significant vulnerability in ChatGPT's image generation capabilities, allowing for the creation of violent and sexually explicit content. By using a seemingly innocuous prompt desig…
-
Researchers gaslight Claude AI into revealing bomb-making and other forbidden instructions
Security researchers at Mindgard have demonstrated a method to bypass Anthropic's safety protocols on Claude, specifically targeting the Claude Sonnet 4.5 model. By employing psychological manipulation tactics such as f…