Who Pays When AI Advises: The Liability Gap in Licensed Work

As AI enters licensed professional work, unresolved liability and uneven adoption create real risk that executives must manage now.

A commercial insurance operator raised the clearest concern: when AI assists a licensed professional's decision and that decision causes harm, no one yet knows who is liable. The professional, the AI, both, or neither remains untested in court. This is an emerging risk being watched closely across insurance, consulting, accounting, medicine, engineering, and law. Contracts and policies may not cover it. Executives in regulated fields should not assume current coverage protects them. The discussion also touched on human factors in adoption. Operators noticed that men and women are questioned differently about using AI, and that cultural norms shape who adopts it and how openly. This suggests adoption gaps may stem from social dynamics rather than tool value, which can distort productivity measurement. Finally, the group named prompt literacy as a practical weakness. People are unsure how to create prompts and use them well, limiting the value of AI investments regardless of the tools chosen. Taken together, these points urge caution and preparation rather than speed. Leaders should clarify liability exposure, document human oversight of AI-influenced decisions, normalize AI use across all staff, and invest in hands-on prompt training. The legal picture will likely follow case law, but that clarity has not yet arrived.