Ask HN: Has your upper management been one-shotted by AI hype?
3 comments
Yes.
The C-suite (and the Board) jumped on to it as a way to reduce headcount and cost, and have stuck their heads in the sand when data shows project velocity is slowing, and bugs are getting created more frequently. Primarily because they have attempted to replace experienced teams with Cursor and cheap outsourcing.
The allure is apparently too strong to face reality that AI can be a powerful addition to the toolbox for productivity, but it is not a replacement for actual skill and experience.
The C-suite (and the Board) jumped on to it as a way to reduce headcount and cost, and have stuck their heads in the sand when data shows project velocity is slowing, and bugs are getting created more frequently. Primarily because they have attempted to replace experienced teams with Cursor and cheap outsourcing.
The allure is apparently too strong to face reality that AI can be a powerful addition to the toolbox for productivity, but it is not a replacement for actual skill and experience.
The problem is management confusing "AI can process text" with "AI can replace judgment." AI is great when the task is bounded, like summarizing docs or triaging info. Understanding why a legacy system was built a certain way requires context that lives in people's heads, not in the codebase.
Some examples:
Engineer says there's a problem or something won't work, and manager+ role uses Cursor to double check there work and proposes some slop solution that AI comes up with. This wastes engineers time by having to sift through this for no reason. It also erodes trust that engineers feel from higher ups, who convey that they who haven't coded + Cursor can do a better job than people coding in the trenches day in and day out.
Other example, engineers say they need a few days to dig into an old codebase/repo to understand some things. Higher up says you can just use Cursor and understand it in an hour(which is never the case since LLMs have no tribal/business context on things).
Wondering if others have seen or experienced the same thing. I think this would make for really bad engineering cultures for teams