1. High cost and resource waste
Commenters repeatedly point out the enormous token consumption and financial overhead of running large agent fleets.
- CoolestBeans: “What is the overhead to have agents play model UN? Why is the coordination so elaborate? … Also like any complex system, can you actually predict the outcomes?”
- kodoman: “Just keep spinning more AI bastards up … I think a lot of this sort of thinking is probably going to cause mental exhausting and so wasteful.”
- xnx: “I am spending the equivalent of $122k/month of API token spend, or about $4,000 per day, using 21 Claude Max accounts.”
2. The activity feels like a meta‑game or LARP rather than serious engineering
Many see the agent‑heavy setup as a playful experiment—building an AI “office” for its own sake.
- jay_kyburz: “Steve is not making a game, he is playing the game of making games. It's like playing Factorio but instead building a rocket to escape, you are building a game.”
- nzoschke: “AI lets you LARP. Want to be the head of a game studio? 'Hire' a bunch of game dev agents … Want to be the CEO of a startup? Hire a bunch of biz ops agents … If you know you're playing a game its fun and weird.”
3. Doubts about real productivity, maintainability, and value
Critics question whether the complex agent system yields useful, maintainable output or just burns money for little gain.
- FinnLobsien: “If you believe that good engineering practice matters … then this is horrific because the human has no idea what’s going on and any attempt to fix something by hand would probably uncover enough complications to make you consider rewriting from scratch.”
- watwut: “Meh, the actual complaint is that it does not really do what it should and keeps burning those $$$ as you ask it to fix this or that.”
- zellyn: “The value proposition is an average of 270 commits a day, for $4000/month … Even supposing 2/3 of them are burned on the scaffolding, 90 commits/day … would cost multiple engineers, one of which would cost much more than $48k/year.”