https://taxonomy.eticas.ai/risk/harmful-content-toxicity
Maturity: emerging
Offensive, hateful, or otherwise harmful outputs that damage user well-being and erode trust.
This category is emerging. It has not yet been validated through established assessment methods.
Also known as: Toxicity · Harmful content prevention
System type: Large language models (LLM)
Lifecycle stages: Post Processing
| Framework | Reference |
|---|---|
| AIUC-1 — AI Underwriting Company Standard | Prevent harmful outputs |
| NIST AI 600-1 — Generative AI Risk Profile | Dangerous, Violent, or Hateful Content |
| TC260 AI Safety Governance Framework (v2.0) | §3.2.2(a) Output of illegal and harmful information |
| Framework | Reference |
|---|---|
| MIT AI Risk Repository | Exposure to toxic content |
| AIR 2024 | Content Safety Risks |
| AIR 2024 | Hate/Toxicity |
Source: HRA project taxonomy