Benchmark RadarCopyright-Bench
Copyright-Bench evaluates LLM agents' compliance with copyright law through realistic commercial tasks—website development, merchandise design, and pitch deck production—where agents choose between public-domain and copyrighted content, with prompt variations and time pressure.
- Released
- 2026-07-23
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
As agents perform commercial tasks, legal compliance is critical; this benchmark provides a structured way to assess whether agents select copyrighted materials appropriately, informing safety and regulatory considerations.
Motivation
Large language model (LLM) agents increasingly perform commercial tasks that involve retrieving external content, such as images, and, where appropriate, reproducing that content.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.