Or that those “nonsense” phrases are not actually nonsense when spoken by a manager. The conclusion they’re nonsense comes from the random generation and the t

Or that those “nonsense” phrases are not actually nonsense when spoken by a manager. The conclusion they’re nonsense comes from the random generation and the technical perspective on semantics; but it’s entirely possible they’re generating phrases that do have semantic meaning when said by a manager… and hence their whole study is flawed. They quietly assume their conclusion, when assuming their generated phrases are vacuous rather than contain coded semantic content.