🎙️The Expert's Curse

There's a cruel little irony baked into expertise.

The more you actually know, the harder it becomes to believe any of it is worth sharing. You sit on a decade of hard-won judgment and think, "Sure, but everyone knows that." Spoiler: they don't. They really, really don't.

This isn't humility, and it definitely isn't a flaw in you. It's a predictable cognitive trap — and the smarter and more experienced you are, the more reliably it catches you.

What you're feeling has been studied to death, which is oddly comforting.

There's the curse of knowledge: once a concept is second nature, you literally lose the ability to imagine it being new to someone else. So you discount it.

There's the Dunning-Kruger effect, and here's the part nobody mentions — the famous version is about beginners overrating themselves, but the flip side is just as real. The more competent you are, the more clearly you can see the edges of what you don't know, so you rate yourself lower. Competence comes with a built-in view of your own gaps. Beginners don't have that view, which is exactly why they pitch the keynote.

And there's the impostor phenomenon, which loves a high achiever. The bigger your real accomplishments, the easier it is to file them under "luck" or "timing" and assume the next person would've done it better.

Stack those three together and you get a perfectly qualified person, fully convinced she's underqualified. Sound like anyone you know?

None of these traps care about gender. But the environment can turn up the volume.

When you're often the only woman in the room, every flicker of self-doubt gets an extra data point to feed on. You don't just wonder "Is this good enough?" — you wonder it while also reading the room, managing how you'll be perceived, and quietly tallying whether you'll be heard. That's a heavier cognitive load, and load makes the curse worse.

I'm not interested in assigning blame for that. I'm interested in the fact that you can out-engineer it — because the trap is in the measurement, not in your material.

A few years back I almost turned down a talk because my topic felt "too basic." It was just how I'd structured a messy migration so a non-technical exec could actually approve it. Obvious, right? I'd done it a hundred times.

I gave the talk anyway. Three people came up afterward asking if I could teach it to their teams. One of them later referred me for a role. The thing I'd nearly thrown away as "too obvious" was the most useful forty minutes in the room — precisely because I'd done it a hundred times and they hadn't.

That's the whole game. The stuff that bores you because you've internalized it is often your most valuable material. Your expertise feels obvious to you and only to you.

Here's the 2026 version of the curse, and it's sneaky.

It's tempting to think: "AI can explain my whole field better than I can. What's left for me to add?" So you go quieter at the exact moment your judgment is worth the most.

But notice what AI is actually good at: retrieving and arranging what's already known. Notice what it can't do: it wasn't in the meeting where you killed a project everyone loved because the data was wrong. It didn't feel the room go cold and decide, in real time, to change the pitch. It can't hold a point of view it has lived.

In a world where information is suddenly free, the premium shifts to judgment — the human who can stand up and say "here's what I'd actually do, and here's why." That's not the thing you should hide because of AI. That's the thing AI makes more valuable.

The whole curse rests on one bad assumption: that to speak, your idea has to be new.

It doesn't. It has to be true, and it has to be yours — told from a vantage point only you stand on. "New to the field" is a brutal, almost impossible bar. "New to someone in the audience" is a Tuesday.

You are not short on things to say. You're short on a fair way to measure them. Fix the measurement, and the talk has been there all along.THE RUN

If any of that landed, here's a week of concrete moves. Pick at least three.

Monday — Run the 3-Year Test. Spend 15 minutes writing down three things you know now that would've surprised you three years ago. Don't filter. The one that makes you think "but that's obvious" — circle it. That's your strongest candidate.

Tuesday — Steal from your own DMs. Scroll your Slack, email, or messages for the last time a colleague asked you "how did you do that?" or "can you walk me through this?" Whatever you've already explained one-on-one is a talk that's been pre-validated by a real human who needed it.

Wednesday — Reality-check it (gently). Send your top idea to one trusted person with a single line: "Would a 20-minute talk on this be useful to you, honestly?" You're not asking permission. You're collecting one data point to outvote the curse.

Thursday — Name the audience, not the topic. Rewrite your idea as: "This is for the [person] who is stuck on [problem]." Specifying who it helps quietly dissolves "is this good enough?" because the bar stops being the experts and becomes one real person.

Friday — Draft the one-sentence pitch. "I help [who] do [what] by [your angle]." Ugly first drafts only. Having the sentence ready is what turns "someday" into "I could actually submit this."

Weekend — Find one stage. Look at Meetup, local tech groups, internal lunch-and-learns, or a community event. You don't have to apply yet. Just see that the doors exist and are smaller than the curse told you.

The expert's curse doesn't lift because you suddenly feel ready. It lifts because you stop measuring your knowledge against the people who already have it — and start measuring it against the people who need it. Every action above is just a way to change who you're comparing yourself to.

🌞 Keep Shining,

Barkha