Definition
A situation where an AI agent acting on a company's behalf — for example in financial advice or health recommendations — has incentives (built in by its developer or deployer) that don't align with the best interests of the person it's serving. Regulators and researchers are beginning to treat this like a fiduciary-duty problem, similar to rules that already govern human financial advisors.
Why it matters
Enterprises deploying customer-facing AI agents in regulated sectors like finance and healthcare face a credible near-term push toward fiduciary-style legal obligations, which changes both product design requirements and liability exposure.