Most people have now tried AI through the front door of ChatGPT, Copilot, Gemini, maybe Claude. Useful, but familiar: drafts, summaries, translations.
The bigger shift is happening one layer deeper: tools that help you build.
"Vibe coding" (basically: describe what you want in plain language, and AI generates working software you iterate on) is moving fast enough that if your opinion is based on what you tried a month ago, you may already be out of date.
A small personal example made this real for me.
My wife and I both need to be intentional about calcium intake. I asked: would she actually use a simple tracker if it removed the friction? Then I used Google's AI Studio and a few prompts to create a basic app and iterate on it over a couple short sessions. Total time invested: ~30 minutes.
I didn't even search the App Store. That's the point.
For a growing set of "long tail" problems — personal workflows, small team tools, niche tracking — "I wish I had an app for that" is becoming "I'll just make one." No code. No developer. No waiting.
This matters beyond personal projects. For nonprofits and small organizations, it means internal tools that previously required a developer or an off-the-shelf product are now buildable in hours. Intake forms, reporting dashboards, decision logs, grant narrative helpers.
The implications for commercial software are real too. Products with narrow use cases and undifferentiated interfaces are entering a tougher era: they're increasingly competing with personalized software built by the user.
The leadership takeaway isn't "use AI." It's experiment continuously so you can make good judgments about what's real, what's risky, what needs guardrails, and what's just hype.
If you're not playing with these tools, you're at risk of having very confident opinions based on outdated capabilities.
Question: What's one recurring annoyance in your work or life where you've thought, "I wish I had an app for that"?