all InfoSec news
ChatGPT Custom Instructions: Persistent Data Exfiltration Demo
July 24, 2023, 2:26 p.m. |
Embrace The Red embracethered.com
As more features are added to ChatGPT the exfiltration angle becomes more likely to be abused.
Recently OpenAI added Custom Instructions, which allow to have ChatGPT always automatically append instructions to every message exchange.
An adversary can abuse this feature to install a data exfiltration backdoor that depends on, and only works because of the image markdown injection vulnerability.
abuse adversary chatgpt data data exfiltration demo exchange exfiltration features image message openai persistent vulnerable
More from embracethered.com / Embrace The Red
Bobby Tables but with LLM Apps - Google NotebookML Data Exfiltration
2 weeks, 2 days ago |
embracethered.com
HackSpaceCon 2024: Short Trip Report, Slides and Rocket Launch
2 weeks, 4 days ago |
embracethered.com
ASCII Smuggler - Improvements
1 month, 3 weeks ago |
embracethered.com
ChatGPT: Lack of Isolation between Code Interpreter sessions of GPTs
2 months, 2 weeks ago |
embracethered.com
Video: ASCII Smuggling and Hidden Prompt Instructions
2 months, 2 weeks ago |
embracethered.com
Jobs in InfoSec / Cybersecurity
Social Engineer For Reverse Engineering Exploit Study
@ Independent study | Remote
Security Engineer II- Full stack Java with React
@ JPMorgan Chase & Co. | Hyderabad, Telangana, India
Cybersecurity SecOps
@ GFT Technologies | Mexico City, MX, 11850
Senior Information Security Advisor
@ Sun Life | Sun Life Toronto One York
Contract Special Security Officer (CSSO) - Top Secret Clearance
@ SpaceX | Hawthorne, CA
Early Career Cyber Security Operations Center (SOC) Analyst
@ State Street | Quincy, Massachusetts