The Rise of the “Confidently Wrong” Colleague (Now Powered by AI)

There’s always one.

You ask a simple question… and they immediately have an answer. No pause. No doubt. Just confidence.

The only problem?
It’s completely wrong.

Turns out, AI is now that colleague.

Recent studies show newer AI models are actually more likely to give a wrong answer than admit they don’t know. Why? Because they’re trained to guess rather than say “I don’t know”.

Sound familiar?

Early in your career, it’s tempting to do the same. To fill the silence. To look smart. To have something to say.

But here’s the truth the best operators learn early:

“I don’t know” is not weakness. It’s credibility.

What does matter:

• Knowing how to find the answer
• Asking better follow-up questions
• Saying**“I’ll come back to you”  – and actually doing it**
• Being curious over performative

Because the workplace doesn’t reward people who guess fastest.
It rewards people who learn fastest.

The irony?
The colleague (or AI) who always answers loses trust over time.

The one who pauses, thinks, and occasionally says “I don’t know”…
becomes the person everyone trusts when it really counts.

Be that person.