Generative AI Model Alignment - AI Mitigation
AI MitigationWhen training or fine-tuning a generative AI model it is important to utilize techniques that improve model alignment with safety, security, and content policies. The fine-tuning process can potentially remove built-in safety mechanisms in a generative AI model, but utilizing techniques such as Supervised Fine-Tuning, Reinforcement Learning from Human Feedback or AI Feedback, and Targeted Safety Context Distillati...
Overview
A source-backed snapshot of this defense.
When training or fine-tuning a generative AI model it is important to utilize techniques that improve model alignment with safety, security, and content policies.
The fine-tuning process can potentially remove built-in safety mechanisms in a generative AI model, but utilizing techniques such as Supervised Fine-Tuning, Reinforcement Learning from Human Feedback or AI Feedback, and Targeted Safety Context Distillation can improve the safety and alignment of the model.
Safeguard details
Where this defense applies and how the source classifies it.
- ATLAS ID
- AML.M0022
- Priority score
- 35
Covered techniques
Attacks this defense is designed to help with.
AML.T0053 - AI Agent Tool Invocation
Model alignment can improve the parametric safety of a model by guiding it away from unsafe prompts and responses.
AML.T0062 - Discover LLM Hallucinations
Model alignment can help steer the model away from hallucinated content.
AML.T0056 - Extract LLM System Prompt
Model alignment can improve the parametric safety of a model by guiding it away from unsafe prompts and responses.
AML.T0057 - LLM Data Leakage
Model alignment can improve the parametric safety of a model by guiding it away from unsafe prompts and responses.
Showing 4 of 7
Source evidence
Original public records and references for this page.
Original source
Original source links
Open the public records and source datasets used for this page.