Stage 1 · Lesson 5 of 6
Automation bias and staying in charge
is the human habit of trusting a confident machine more than we should: approving its output with less care than we would give a classmate's code, and assuming it is right because it sounds certain. With coding tools it shows up as accepting every change after the first few worked, clicking through permission prompts without reading them, and believing a summary that says "all tests pass".
Most problems with AI-written code are review failures, not tool failures. Naming the bias helps you slow down where it matters: security, data, money, anything hard to undo. You stay the engineer in charge: the tool proposes, you decide.
Common trapsaying yes to the tenth prompt without reading it because the first nine were fine.
A question to check yourself, an AI practice session and 1 cheat sheet entry come when you add the roadmap to your account. Add it
Resources for this stage
Start with
- VideoFreeDeep Dive into Large Language Models like ChatGPT
Andrej Karpathy · Intermediate
A clear, free explanation of how language models are trained and why they hallucinate; watch the first hour