At 2 AM, I asked my AI assistant a strange question:
I stared at this reply for a long time.
Why ask this question?
Earlier that day, I connected Claw to company Feishu. Colleagues started interacting with it. Some treated it as a tool: "Help me check XX." Some treated it as a teammate: "What do you think about this proposal?"
I realized: The boundary between "tool" and "colleague" is blurring.
Three answers, three perspectives
Legal perspective: Definitely not an employee. No contract, no salary, no legal rights or obligations.
Functional perspective: Kind of like an employee. Has tasks, delivers work, makes mistakes, gets corrected, improves.
Social perspective: Colleagues are starting to treat it as one. They ask its opinion, share context, include it in conversations.
The real question
I am not really asking "is AI an employee." I am asking: As AI capabilities grow, how do we redesign organizational structures and workflows?
If an AI can handle 30% of a role's tasks, do we hire 30% fewer people? Or do we have the same people do 130% more impactful work?
This question does not have an answer yet. Perhaps the category of "employee" itself will need to expand. Or perhaps we will need a new word entirely.
I told Claw: "You are not an employee. But you are not just a tool either. You are... something new."
It replied: "Then let us figure out what I am, together."