What copyright / IP issues arise with LLMs in production?

hard

Answer

$(1) Training data: many LLMs are trained on copyrighted works (books, code, images) — ongoing lawsuits (NYT v OpenAI, Getty v Stability).$
$(2) Output: models can regurgitate copyrighted text verbatim (rare but possible).$
$(3) Code: models can output code covered by copyleft licenses (GPL) without attribution — Copilot lawsuit.$
$Mitigations: attribute sources where possible, offer indemnification (OpenAI, Microsoft, Anthropic do for enterprise), keep an audit trail of prompts/outputs, and consult legal for high-risk domains (music, film).$

Check yourself — multiple choice

No copyright risk
Training data (lawsuits), verbatim regurgitation, license-tainted code — mitigate via attribution, indemnification, audit logs, legal review
Only for images
Solved by open source

LLM IP: training suits + regurgitation + license leakage; mitigate via attribution + indemnification + audit.

#safety#production#reliability

214 interview questions in this topic.