Notify users before high-impact or generative AI service use, and label generative or realistic synthetic outputs.
Legal basis: Act Art. 31(1), Act Art. 31(2), Act Art. 31(3), Decree Art. 23
AI Basic Act readiness
Can you explain the evidence behind each AI answer?
Readiness is not only a disclosure banner. You need to explain which model used which approved business definition and context version at answer time.
Three axes
Readiness splits into transparency, high-impact AI management, and safety evidence.
rawctx pins business meaning by package, version, and context hash, then records which answer used which reference through Answer Audit. This self-check finds where that evidence layer matters.
Pre-check statutory high-impact AI areas and keep explanation, user-protection, human-oversight, and documentation controls.
Legal basis: Act Art. 2(4), Act Art. 33(1), Act Art. 34(1), Decree Arts. 25 and 27
Keep risk management, incident response, explanation evidence, and implementation documents for large-scale and high-impact AI.
Legal basis: Act Art. 32(1)-(2), Act Art. 34(1)1 and 5, Decree Arts. 24 and 27(2)
AI Basic Act
Self-diagnosis
Check readiness from the way your AI feature is used and the evidence your operation keeps.
This tool is not legal advice or a compliance determination. It organizes evidence and review priorities for AI Basic Act readiness.
Question 1/24Transparency
Generative AI is involved in answers, summaries, recommendations, reports, or documents users see.
If the system mimics input-data structure and characteristics to generate text, sound, images, video, or similar outputs, review the generative AI duties.
Legal basis: 법 제2조제5호, 법 제31조제1항, 법 제31조제2항
Notify users before providing high-impact or generative AI services, and label generative or realistic synthetic media outputs.
Legal basis: 법 제31조제1항, 법 제31조제2항, 법 제31조제3항, 시행령 제23조제1항부터 제3항까지