Workplace policies about data handling, internal communication and AI disclosure read like corporate subjects, and they are usually taught as though nobody had ever encountered the underlying question before. They have. Whether to forward the screenshot, whether to repeat the thing you cannot verify, whether to mention how you produced the work — these are the same three decisions, first met around thirteen.
Same Habits, Different Stakes
The three areas do not change between a group chat and a company: information about people, claims you cannot verify, and how you treat the people involved. What changes is who is standing underneath when something is dropped.
At school the cost lands on classmates, and it is real. At work it can land on a customer who was never in the conversation, on a colleague whose situation you repeated, or on the company. The skill did not get harder. The blast radius did.
Data You Touch That Is Not Yours
Almost every job now involves handling information about people who are not in the room. A support ticket, a customer record, a spreadsheet with a hundred names in it, a screenshot of an account taken to ask a colleague a question.
The reflex that matters is the one from the permission screen: what is this, who else will end up seeing it, and how long will it exist? Pasting a customer record into a public channel to ask a quick question is the workplace version of uploading a contact list — done casually, done to be helpful, and made of other people.
The Forwarded Claim, at Work
Internal misinformation is rarely dramatic. It is "I heard the client is unhappy", passed on twice, arriving at a manager as established fact. Nobody lied. Every step was somebody helpfully sharing what they had been told.
The check is the same one that works on a video: what would have to be true, and who would know? The extra move that costs almost nothing is saying where it came from. "Someone mentioned the client might be unhappy — I have not confirmed it" carries the same information without the false certainty.
AI at Work: The Same Disclosure Question
Workplace debates about AI tools rerun the schoolwork argument almost line for line, and land in the same place. The question is rarely whether the tool is permitted. It is whether the person relying on the output knows how it was made, and whether the person who produced it can stand behind it.
The check also survives the move: could you explain any part of what you are handing over, and answer a question about it? If not, you are passing along something you have not verified, and the fact that a tool wrote it does not transfer the responsibility to the tool.
Where This Is Worth Starting
None of this argues for teaching workplace policy to twelve-year-olds. It argues the other way: the habits form early, in low-stakes settings, and adding the workplace consequences later is much easier than installing the habit for the first time in someone who has spent a decade forwarding things without checking.
The Digital Ethics & Citizenship test is deliberately built for grades 6-8 and stays there — its scenarios are school scenarios and it is not a hiring instrument. For adults, Work Ethics & Reliability and Situational Judgement ask workplace versions of the same underlying questions.