Skip to player
Research
Offensive AI Security
›
Alignment & Fine-Tuning Attacks
›
Harmful Fine-Tuning: Dose-Response and the Capability-Safety Decoupling
Contents
Host
Expert
Murali Chillakuru
Press play to begin the walkthrough.
0:00
16:45
1×
1.25×
1.5×
0.85×
CC
Alignment & Fine-Tuning Attacks · 2 / 5
Harmful Fine-Tuning: Dose-Response and the Capability-Safety Decoupling
Play
Back