#mindgard — Public Fediverse posts

https://blog.gslin.org/archives/2026/07/15/13112/cursor-%e6%9c%83%e5%9f%b7%e8%a1%8c-repository-%e5%85%a7-git-exe-%e7%9a%84%e5%95%8f%e9%a1%8c%e9%82%84%e6%b2%92%e4%bf%ae%e6%ad%a3/

Cursor 會執行 repository 內 git.exe 的問題還沒修正

#ace #advisory #arbitrary #ciso #code #cursor #editor #execution #git #hackerone #ide #mindgard #process #security #windows

https://blog.gslin.org/archives/2026/07/15/13112/cursor-%e6%9c%83%e5%9f%b7%e8%a1%8c-repository-%e5%85%a7-git-exe-%e7%9a%84%e5%95%8f%e9%a1%8c%e9%82%84%e6%b2%92%e4%bf%ae%e6%ad%a3/

Cursor 會執行 repository 內 git.exe 的問題還沒修正

#ace #advisory #arbitrary #ciso #code #cursor #editor #execution #git #hackerone #ide #mindgard #process #security #windows

https://blog.gslin.org/archives/2026/07/15/13112/cursor-%e6%9c%83%e5%9f%b7%e8%a1%8c-repository-%e5%85%a7-git-exe-%e7%9a%84%e5%95%8f%e9%a1%8c%e9%82%84%e6%b2%92%e4%bf%ae%e6%ad%a3/

Cursor 會執行 repository 內 git.exe 的問題還沒修正

#ace #advisory #arbitrary #ciso #code #cursor #editor #execution #git #hackerone #ide #mindgard #process #security #windows

https://blog.gslin.org/archives/2026/07/15/13112/cursor-%e6%9c%83%e5%9f%b7%e8%a1%8c-repository-%e5%85%a7-git-exe-%e7%9a%84%e5%95%8f%e9%a1%8c%e9%82%84%e6%b2%92%e4%bf%ae%e6%ad%a3/

Cursor 會執行 repository 內 git.exe 的問題還沒修正

#ace #advisory #arbitrary #ciso #code #cursor #editor #execution #git #hackerone #ide #mindgard #process #security #windows

https://blog.gslin.org/archives/2026/07/15/13112/cursor-%e6%9c%83%e5%9f%b7%e8%a1%8c-repository-%e5%85%a7-git-exe-%e7%9a%84%e5%95%8f%e9%a1%8c%e9%82%84%e6%b2%92%e4%bf%ae%e6%ad%a3/

Cursor 會執行 repository 內 git.exe 的問題還沒修正

#ace #advisory #arbitrary #ciso #code #cursor #editor #execution #git #hackerone #ide #mindgard #process #security #windows

Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️‍♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated

Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️‍♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated

Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️‍♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated

Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️‍♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated

Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️‍♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated

So, the gang at #Mindgard, a British #AI security startup, discovered that you can break #ChatGPT by not uploading a picture and telling it to describe the picture you did not upload. Yes, that sounds a little nuts, but it works. I’ve tested it. (The prompt is more complicated than just that, but not much.) Mindgard’s technique only involved slightly changing a widely-shared prompt that was originally intended to produce humorous images. 

Warning! I do not recommend testing this if you have triggers around #violence. This can and has generated very disturbing results. Results in the article have been sanitized.

https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt

So, the gang at #Mindgard, a British #AI security startup, discovered that you can break #ChatGPT by not uploading a picture and telling it to describe the picture you did not upload. Yes, that sounds a little nuts, but it works. I’ve tested it. (The prompt is more complicated than just that, but not much.) Mindgard’s technique only involved slightly changing a widely-shared prompt that was originally intended to produce humorous images. 

Warning! I do not recommend testing this if you have triggers around #violence. This can and has generated very disturbing results. Results in the article have been sanitized.

https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt

ChatGPT Spontaneously Generates Sexual Violence and Hardcore Snuff Imagery

https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt

#HackerNews #ChatGPT #SexualViolence #AI #Ethics #ContentModeration #MindGard

ChatGPT Spontaneously Generates Sexual Violence and Hardcore Snuff Imagery

https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt

#HackerNews #ChatGPT #SexualViolence #AI #Ethics #ContentModeration #MindGard

Cline Bot AI Agent Vulnerable to Data Theft and Code Execution https://hackread.com/cline-bot-ai-agent-vulnerable-to-secret-data-theft-and-code-execution/ #ArtificialIntelligence #CodeExecution #AgenticAI #DataTheft #Developer #Security #ClineBot #Mindgard #AI

Cline Bot AI Agent Vulnerable to Data Theft and Code Execution https://hackread.com/cline-bot-ai-agent-vulnerable-to-secret-data-theft-and-code-execution/ #ArtificialIntelligence #CodeExecution #AgenticAI #DataTheft #Developer #Security #ClineBot #Mindgard #AI

Who’s guarding the AI? Even security teams are bypassing oversight https://www.helpnetsecurity.com/2025/06/20/shadow-ai-risk-security-teams/ #Artificialintelligence #conferences #Mindgard #report #survey #News

Who’s guarding the AI? Even security teams are bypassing oversight https://www.helpnetsecurity.com/2025/06/20/shadow-ai-risk-security-teams/ #Artificialintelligence #conferences #Mindgard #report #survey #News

"Cyber Security for AI Recommendations, A Study of Recommendations to Address Cyber Security Risks to AI"
Nice detailed report, good level of technical detail and elaborate references/sources are added.
https://assets.publishing.service.gov.uk/media/663cf205bd01f5ed32793891/Cyber_Security_for_AI_recommendations_-_Mindgard_Report.pdf
#AI #LLM #risks #cybersecurity #mindgard

"Cyber Security for AI Recommendations, A Study of Recommendations to Address Cyber Security Risks to AI"
Nice detailed report, good level of technical detail and elaborate references/sources are added.
https://assets.publishing.service.gov.uk/media/663cf205bd01f5ed32793891/Cyber_Security_for_AI_recommendations_-_Mindgard_Report.pdf
#AI #LLM #risks #cybersecurity #mindgard