I have divided the work into 4 parts 1. Gather enough data which has all the details of attacks and responses to those attacks. 2. Clean and prepare that data for training an open-source model (Add the ability to scan, control, defend on any misbehaviour with the system). 3. Train it and test it in a battelground (I feel this is risky and I need help here). 4. Try to deploy it and improve it, so that people defend themselves easily (it will be open-source).
I will build an LLM to actively defend cyber attacks
- Posted 1 hour ago by mastaanshaik
- 1 points