---
title: "Comment by Abhinav Sharma on \"Code Is Cheap. Authority Isn't.\" | Hashnode"
description: "Review time up 91% while PRs got 154% larger means review is the thing that gives, and once merged-with-no-review becomes normal it stops looking like a decisio"
url: "https://hashnode.com/posts/code-is-cheap-authority-isn-t/6ad54358d75f37b1b40b3b8d/comment/6ad56d2839cde4bbff5982c3"
---

[Discussion](https://hashnode.com/posts/code-is-cheap-authority-isn-t/6ad54358d75f37b1b40b3b8d)

Comment [Code Is Cheap. Authority Isn't.](https://hashnode.com/posts/code-is-cheap-authority-isn-t/6ad54358d75f37b1b40b3b8d)

[Abhinav Sharma](https://hashnode.com/@abhinavs1920)

[Oct 19](https://hashnode.com/posts/code-is-cheap-authority-isn-t/6ad54358d75f37b1b40b3b8d/comment/6ad56d2839cde4bbff5982c3)

Review time up 91% while PRs got 154% larger means review is the thing that gives, and once merged-with-no-review becomes normal it stops looking like a decision anyone made. The piece I'd push on: if evidence is the scarce resource, what stops the agent from generating the evidence too? Then the human is grading work against a rubric the worker wrote.

[Arunima Chaudhuri](https://hashnode.com/@ArunimaChaudhuri)

[Nov 3](https://hashnode.com/posts/code-is-cheap-authority-isn-t/6ad54358d75f37b1b40b3b8d/comment/6ad56d2839cde4bbff5982c3#reply-6ae9877eb09e0c496f5ea55e)

You’re right that once merged-with-no-review becomes normal, it stops looking like a decision. That’s why I’d track the unreviewed-merge rate explicitly. It turns a silent default back into a number someone has to answer for.

On your question, I’d separate evidence from the rubric. Agents can and should generate evidence, because that’s the cheap part. What they can’t be allowed to do is define what counts as passing. In practice that means three things:

1.  A human writes the acceptance criteria before the work starts, and they sit somewhere the building agent can’t edit.
2.  The verifier runs in a different session from the builder, with different context.
3.  You check the tests themselves, for example with mutation testing: if you deliberately break the code and the tests still pass, the evidence is theatre.

The risk you’re describing is real when one agent writes the code, the tests and the definition of done. Split those three apart and the human is grading against their own rubric again.
