@Fireship
  @Fireship
Fireship | Rabbit R1 makes catastrophic rookie programming mistake @Fireship | Uploaded 2 months ago | Updated 2 days ago
A group of jailbreakers recently discovered that the Rabbit R1 codebase contains hardcoded API keys - giving them easy access to user data from their AI tech-to-speech service.

#programming #hacking #thecodereport

πŸ”₯ New Full Linux Course coming soon!

Upgrade at https://fireship.io/pro
Use code LINUX30 for 30% off PRO access

πŸ’¬ Chat with Me on Discord

https://discord.gg/fireship

πŸ”— Resources

Rabbitude Disclosure https://rabbitu.de/articles/security-disclosure-1
AI failures https://youtu.be/krixaEhLnlA
CES 2024 highlights https://youtu.be/eHzoTLwx01E

🎨 My Editor Settings

- Atom One Dark
- vscode-icons
- Fira Code Font

πŸ”– Topics Covered

- Problems with Rabbit R1
- What to when sensitive data is leaked?
- Major mistakes made by programmers
- AI tools that have failed
- Hacking incidents of 2024
- API key safety tips
Rabbit R1 makes catastrophic rookie programming mistakeClaude 3 just destroyed GPT-4 and Gemini... AGI is near?Nvidia CUDA in 100 SecondsAnother glorious battle for AI dominance… GPT-4o vs Google I/O100+ Docker Concepts you Need to KnowUnhinged ransomware attack targets hospitalsOpenAI shocks the world yet again… Sora first lookGoogle actually beat GPT-4 this time? Gemini Ultra releasedExpo in 100 SecondsApple drops ban hammer on Epic Games over mean tweetGoogle has the best AI now, but theres a problem...

Rabbit R1 makes catastrophic rookie programming mistake @Fireship