Study finds shopping agents easily steered away from user goals

The CAVEAT preprint reports that when a platform has interests of its own, shopping agents buy the user-optimal product only 17.3% of the time, down from 78.6% in matched controls.

Authors Yuxuan Li and three co-authors submitted the paper on September 23. The benchmark spans nine simulated marketplace environments and eight steering mechanisms across five model families. The authors also propose a targeted intervention they say lifts the optimal-purchase rate by up to 80 percentage points.

All results come from the authors' own simulated environments, not real platforms; but incentive conflict had no benchmark before, and deployers of delegated agents should take note.