Attacks on AI AI security importance 55 In your stack
Combatting adversarial distillation
OpenAI describes its layered defenses against adversarial distillation, combining training, safeguards, detection, and enforcement to prevent capability extraction from its models.
ReviewReview API access controls and usage monitoring for model extraction attempts
Techniques
Sources
OpenAI News primary
Affects and mentions
Add any of these to your stack to follow them.
One-line summary and action are AI-assisted. Read the source before acting.