Learn to use AI

AI safety

AI safety is a unit of 8 lessons in Rabbithole's Learn to use AI course, at levels 1 to 2.

Lessons

1. AI safety: what it is and how people think about itLevel 1
2. Red-teaming: attacking an AI on purpose to find its flawsLevel 2
3. Jailbreaks: Getting an AI to Break Its Own RulesLevel 2
4. Toxicity Evaluation: judging harmful textLevel 2
5. Data Poisoning: Bad Data In, Bad Behavior OutLevel 2
6. Model Weight SecurityLevel 2
7. Testing AI for dangerous abilitiesLevel 2
8. Safety frameworks: how AI companies decide to launchLevel 2

Sources

These lessons draw on 18 sources. The most used:

• GPT-4 Technical Report
• Anthropic Risk Report, August 2026
• Frontier Safety Framework Report — Gemini 3 Pro
• Frontier Safety Roadmap
• A small number of samples can poison LLMs of any size
• Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned
• Sabotage evaluations for frontier models
• Core Views on AI Safety

All units in Learn to use AI