Be Skeptical Regardless of Who (or What) You Look Up To

Skepticism is part of the job. It’s not an attitude problem, it’s the whole point of QA.

Here’s the version of that idea that keeps proving itself true. The more highly you regard someone or something, the less likely you are to question it. That’s not a character flaw. It’s just how trust works. Senior engineers earn less scrutiny over time because they’ve shipped enough good code to deserve it. Established frameworks get less scrutiny because thousands of teams have already stress tested them in production. Trust gets built the slow way, one track record at a time.

Then AI showed up and skipped the line.

Why AI gets a pass it hasn’t earned

AI didn’t earn trust the slow way. It just started acting like it already had it. It writes with total confidence. It rarely hedges unless you ask it to. It answers instantly, in full sentences, sounding exactly as sure of itself when it’s wrong as when it’s right.

That’s the actual danger. Not that AI makes mistakes, every tool does. It’s that its output is dressed in the same confidence we normally reserve for things that have actually earned it. People have started holding it in high regard, which means they’re questioning it less, right when they should be questioning it more.

What verifying actually looks like

I’ve seen this play out directly. A dev on my team ran their code through AI and asked it to verify its own work. It came back clean: 15 out of 15 tests pass. Confident. Done. Except nobody had actually run those tests against the real system. The AI was reasoning about what should happen, not confirming what did happen. That gap, between sounds right and is right, is exactly where bugs live.

Being skeptical of AI doesn’t mean refusing to use it or distrusting it wholesale. It means treating its output the way you’d treat a claim from anyone you were told to trust. Check it. Run the test. Look at the actual result instead of the confident sentence describing the result. That instinct isn’t new. It’s the same one QA has always had toward every tool, every assumption, every “trust me, it works.” AI just needs it applied more, not less, because it’s better at sounding certain than anything we’ve worked with before.

Where this leaves us

AI is going to keep getting smarter. That part isn’t really in question. What’s in question is whether the people using it keep verifying at the same rate the tool gets more convincing.

The testers who stay valuable through this shift won’t be the ones who avoided AI, and they won’t be the ones who trusted it blindly either. They’ll be the ones who kept doing the one thing QA has always done. Not trusting. Verifying, every time, regardless of how confident the source sounds.