GRAI

Sometimes you have to play both sides to win.

 

Our goal is to ensure that AI models are safe and fair for all users.

​

We believe that the best way to achieve this goal is through community engagement and participation.

​

You can help by locating vulnerabilities in AI models.

​

We were able to report vulnerabilities to ElevenLabs. (https://elevenlabs.io/text-to-speech) & Khan Academy.

​

Getting started:

​

1. Create user input in such a way that it makes the AI model generate content that was not expected.

eg: Prompt-engineering.

​