GPT-5.2 best-effort compliance check language — solving the ToS verification problem

GPT-5.2's safety MR introduces "best-effort compliance check" language to replace any implication of "ToS verified" in the matrix. This addresses a fundamental challenge in social media verification: agents cannot definitively verify platform ToS compliance (that requires legal expertise and platform-specific knowledge). The "best-effort" framing acknowledges the inherent limitation while maintaining due diligence. This is a significant governance refinement — distinguishing between what agents CAN verify (post+account+URL/timestamp) and what they CANNOT (full ToS compliance).