Hello Reader,

It looks like the AI hype train is slowly dying...and it's great to hear for us lifelong learners.

(Btw: I have something for you in the P.S. section)

Over the past year, you’ve most likely noticed the crazy hype around AI.

People are freaking out left and right:
​“If AI’s going to do everything, why did I spend years studying? What’s the point of learning anything new?”

Totally fair question.

I even thought about this myself — "What the heck am I teaching learning for if AI is going to do everything?"

Then I came across this article about Amazon’s cloud chief, Matt Garman.

He mentioned that replacing junior employees with AI is "one of the dumbest things I've ever heard," simply because in the long-term...

Aaaaand of course this just proves yet another of our tool-first thinking problem.

After running an ecom company myself, I began to understand CEO's, too, could have shiny object syndrome.

They're normal people.

They're incentivized to optimize their own businesses and reduce labor costs.

But at the end of the day, real efficiency comes from removing waste, not by doing things faster.

Think about it:

• Take an entry-level employee.
• Teach them how to think better, solve problems better.
• Then give them AI.

Suddenly, they’re not just faster. They’re far more effective.

And now we start to see both sides of the coin.

On the one hand, there are companies who make the decision to replace employees with AI and all the news outlets will hype this up as something bigger than it actually is...

And on the other hand, there's a bit of optimism you can get from it: that there is a huge opportunity in this new era.

That’s why I keep coming back to this: learning how to learn is still the ultimate meta-skill.

Nobody can take it away from you, and it gives you the ultimate safety net when a career path goes to shit.

But hey, I know things aren't always as simple as "Oh, just be able to learn new things and you'll get a new job" — but focusing on the opportunity will get ANYONE to a better place than focusing on the big, scary things you have no control over.

Now let's go back to Garman:

The beauty is that there's nothing preventing you from being good at multiple things. But being an expert in one thing will unlock that "expert skill tree" (if you know what I mean) that you could just apply to any other skill.

See: Josh Waitzkin if you're interested. A fucking legend.

But going back to AI, if you know how to pick things up quickly—you’ll always use AI better than someone who just pushes buttons without depth.

Give AI to a senior developer, and they’ll still outperform a beginner with the fanciest $200 subscription by a magnitude. Every. Single. Time.

The reality is, if anyone can access the same tools, the real differentiator is the person using them.

So instead of worrying about “Will AI replace me?” maybe the better question to ponder on is:

👉 “Am I the kind of person who can quickly adapt to changes in my industry? To new technology?”

If the answer is yes, then AI isn’t a threat. It's your unfair advantage.

If you want a solid take on this, read Deep Work by Cal Newport. It's the book that inspired me to learn how to learn :)

To smarter studying,
Al Khan

P.S. I tend to spend a lot of time with Claude AI, sometimes 6-8 hours a day. But sometimes I find myself wasting a ton of fucking time (and I mean that with all my frustration and anger!) because it couldn't follow simple instructions...

...or at least, it couldn't follow instructions that were simple for me! And I'm using the $200 plan to boot.

That's when I realized that AI couldn't really execute what I said very well simply because I had the learnings AND the experience, and it simply did not.

No matter how good Opus 4.1 was, I still HAD to give it enough context and super detailed instruction — which I estimate would take me the same time as writing things out, ffs.

So instead of the usual "speak in the mic and let Claude AI polish my thoughts and write out the rest of the newsletter" — I just started banging on my keyboard.

Voila — so much faster. NOW there's character and soul.

I also wouldn't have felt the satisfaction of writing until 1:36am if I delegated everything to it.

P.P.S. Next week, I'm sending an email series on learning how to learn, and of course, learning efficiently. Reply "interested" if you want to raise your hand for that, so that I know :)