This recent HBR article gives plenty of food for thought.

In an experiment conducted with managers responsible for shelf stocking at Tapestry Inc., a renowned luxury fashion retailer, researchers employed software algorithms to provide recommendations on the selection and display of merchandise.

The managers participating in the experiment received two different types of recommendations during the experiment: either recommendations based on a comprehensible rationale, or "black box" recommendations without any explanation.

>>
When the [managers] received a recommendation from an interpretable algorithm, they often overruled it based on their own intuition. But when the same [managers] had a recommendation from a similarly accurate “black box” machine learning algorithm, they were more likely to accept it even though they couldn’t tell what was behind it. Their resulting stocking decisions were 26 percent closer to the recommendation than the average choice.

Why? Because they trusted their own peers who had worked with the programmers to develop the algorithm.

The allocators “knew that people like them—people with their knowledge base and experience—had had input into how and why these recommendations were being made and had tested the performance of the algorithm,” the researchers write. “We call this social-proofing the algorithm.”
>>

This opens up some really interesting possibilities about the potential impact on decision making and - importantly - decision outcomes, depending on when 'humans in the loop' are empowered to override recommendations and what information they are given (or not) to help make those calls and, equally importantly, how engaged they are in the underlying design of the tools used for decision making.

#ai #aidesign #decisionmaking #trust #socialproof #humansatcenter

https://lnkd.in/gUqShvNj