Top salesperson's outside AI use: ban or not? A 400-employee case

The case is presented as a composite drawn from the author's consulting engagements, not a named company. It follows a string of シャドーAI incidents in which employees adopt outside generative AI faster than their employers can set rules. In October, a company handling the Panasonic Group's administrative headquarters functions disclosed that data including personal information on roughly 330,000 employees and others had been uploaded to a generative AI service it had banned; the company attributed this to rule violations by members rather than an external attack.

The author argues that few employees turn to outside AI out of laziness. The more common driver is that the company-provided tool is not good enough for the work they are trying to get done. That framing is what makes the salesperson's case uncomfortable: punishing the most capable AI user may not remove the motive that sent them outside the internal system in the first place.