Risks & Ethics
AI Safety
Also known as: safety
In one line
The field focused on preventing AI systems from causing harm — accidental or deliberate.
What does AI Safety mean?
AI safety spans everything from bias and misinformation to catastrophic risk from advanced systems. Governance, red-teaming, and alignment research all fall under it.
A real-world example
Anthropic's Responsible Scaling Policy defining capability thresholds that trigger extra safeguards.