OpenAI’s Double Trouble: ChatGPT Security Flaws and Internal Hacker Drama Exposed
OpenAI’s Mac ChatGPT app was caught storing chats in plain text, leading to a scramble for encryption. Meanwhile, a whistleblower claims internal security flaws could be ripe for foreign exploitation. Looks like OpenAI’s cybersecurity strategy could use a little AI assistance itself!
Hot Take:
Looks like the ChatGPT Mac app was trying to play peek-a-boo with your conversations. Meanwhile, OpenAI’s data security might be as solid as a wet paper bag. Time to send in the cyber janitors!
Key Points:
- OpenAI’s Mac ChatGPT app was found to store user conversations in plain text.
- Swift developer Pedro José Pereira Vieito discovered the security flaw.
- The app isn’t available on the App Store, bypassing Apple’s sandboxing requirements.
- OpenAI faced a 2023 hack, exposing internal messaging systems.
- Whistleblower Leopold Aschenbrenner raised concerns but was later fired, allegedly for unrelated issues.
Membership Required
You must be a member to access this content.