Lesson 7 of 7
So, which one?
5 min read
Six lessons in, here's the honest answer to 'which AI should I use?' — it depends on you. But that's not a dodge. It's a decision you can make in ten seconds.
Start from what matters to you
There's no universal best, so stop looking for one. Start from your priority instead. Care most about privacy? Run a local open model. Want the smartest answers? A top reasoning model on a paid tier. Free and fast? A light everyday model. Fresh facts? Search-first. Inside your apps? A copilot. Your priority points straight at a shape.
The question isn't 'which AI is best?' — it's 'what matters most to me for this?' Answer that, and the shortlist writes itself.
Then hold it loosely
Whatever you pick, hold it loosely. The models leapfrog every few months, so re-check now and then, keep a small council for different jobs, and lean on the shapes you learned here rather than a brand name. You're not choosing forever — you're choosing well for now, and staying free to change.
You're not marrying a model. Pick well for today, keep a council, and re-decide whenever the field moves — which it will.
Don't over-optimise. For most everyday tasks any current top model is more than good enough — the choice only really matters at the edges (privacy, cost, hard problems, fresh facts). Save your energy for those.
The shape of it
- —There's no universal best — start from your own priority for the task.
- —Each priority (privacy, smarts, free, fresh facts, in-your-apps) points to a shape.
- —Hold your choice loosely: keep a council and re-decide as the field moves.
After all this, what's the soundest way to answer 'which AI should I use?'
Continue in the app
Take the whole Which AI should I use? course — tracked
Get your personalized path, progress and streaks in the app — this lesson and every next one, in order.