5 AI Lessons from My Latest My First Million Appearance

Woo hoo! My fourth appearance on the My First Million podcast is now live:

As always, it was a blast chatting with Sam Parr and Shaan Puri. There was lots to cover on the AI front, and I'm glad I got to address tool calling, vector embeddings, the future of AI teammates, and all the other takes I ramble about in this newsletter.

I've pulled out five key lessons from our conversation that I think will resonate most with builders, entrepreneurs, and anyone trying to navigate this AI transformation.

In particular, I'm breaking down:

• The mindset shift that separates AI winners from losers

• Why daily AI use isn't optional anymore

• How to give your AI superpowers with your own data

• The evolution from AI tools to AI teammates

• The strategic approach to finding AI opportunities

—@dharmesh

Lesson 1: You^AI - The Power Formula

The biggest mistake I see people make is thinking about AI as "you versus AI."

That's misguided. The right mental framework is You^AI -- AI as an exponential amplifier of your capabilities.

AI isn't going to take your job because it's trying to do things that you do. Instead, it's going to unlock things and let you do things that you were never able to do before, which will increase your value, not decrease it.

Think about it: Are you less valuable as a human because you use spell check instead of a dictionary? Because you use Google instead of memorizing encyclopedias? Because you use a calculator instead of doing long division by hand?

Of course not. You're more valuable because these tools amplify your capabilities.

But the catch is that in order for this to be true, you actually have to use it. You have to learn it. You have to experiment with it.

The people who figure this out early (and yes, it's still early) will have a massive advantage over those who don't.

Lesson 2: Use AI Every Single Day (No Exceptions)

Here's my simple rule: Every day, you should be in ChatGPT (or Claude, Gemini or other frontier AI app).

I don't care what your job is, you could be a waitress at a restaurant and you should be using ChatGPT every day to make yourself better at whatever it is you do.

My practical advice: Anytime you're going to sit down at a computer and do something -- research, writing, analysis, whatever -- you should give ChatGPT a shot at it first.

Pretend like you have access to an intern that has a PhD in everything. Give it a crack at solving the problem you're about to spend time on.

What you'll invariably find is that you'll be surprised by the number of times it actually comes up with a helpful response that you never would have expected. That's because it literally has been trained on everything publicly available.

This daily practice is how you develop intuition for what AI can and can't do, which becomes incredibly valuable over time.

Lesson 3: Give Your AI Superpowers with Your Data

Most people are using AI like a really smart Google search. But the real power comes when you give it access to your personal or business data.

Here's the technical reality: LLMs are limited to what they were trained on (the "knowledge cutoff date"). They don't know anything about your specific business, your emails, your documents, or your personal situation.

The hack that's changed everything is something called vector embeddings and RAG (Retrieval-Augmented Generation). Think of it like this: You take all your important documents and put them in a special database. When you ask a question, the system finds the five most relevant documents and gives them to the AI as context.

It's like hiring a brilliant intern who knows everything about everything, but on their first day of work, you hand them exactly the five documents they need to answer your specific question.

The quality of responses when AI has access to rich, relevant data is shocking compared to generic queries.

Lesson 4: Stop Thinking Tools, Start Thinking Teammates

Most of us use AI as tools right now, which is fine. But you need to shift your thinking and start thinking of them as teammates -- like that intern who just got hired.

As agents get better, every company is going to have hybrid teams consisting of what I like to think of as carbon-based life forms and digital AI agents. It's inevitable.

But one of the biggest problems that many people trying agents right now are facing is that we're hiring these agents and expecting them to do magic with no training, no onboarding, no feedback, no one-on-ones, no nothing. Your results are going to be terrible if you treat them like magic boxes instead of new team members.

This creates massive opportunities for new businesses: What does employee training look like for digital workers? What do performance reviews look like for digital workers? How do we do recruiting for digital workers?

We're going to need software, onboarding systems, training programs, and entirely new management approaches for these hybrid teams.

Lesson 5: Find What AI Can "Sort Of" Do Today

Here's the strategic advice that will separate successful builders from everyone else: Find the frontier of what AI is incapable of.

If AI can "sort of" do something right now, meaning you have to squint a little bit and think "well, it's kind of something, but not quite there" -- wait 6 months or a year.

That's the beauty of an exponential curve. It gets so much better so fast that if it can sort of do it now, it will be able to do it really well soon.

This is actually the smart approach for startups too: Build the thing that AI kind of can't do right now. You just have to stay alive long enough to give it the 12-18 month runway it needs to go from "eh, didn't really work very well" to "oh my god, this is amazing."

But you've built your brand, your company, your customer base along the way. You're basically betting you'll be able to surf the improvement of the models.

The AI Revolution Is Personal

The companies and individuals who adopt the "You^AI" mindset, use AI daily, give it access to their data, treat it as a teammate, and strategically position themselves at the frontier will have advantages that compound over time.

This transformation is happening faster than most people realize. The question isn't whether AI will change how we work -- it's whether you'll be leading that change or reacting to it.

If you want to watch the full conversation with Sam and Shaan, you can check it out on the My First Million podcast. We covered a lot more ground, including tool calling, the future of creativity, and why I think we're creating a new species of human (I cringe a little every time I say that).

Hope you enjoyed these insights. All feedback is welcome, feel free to let me know with the poll below.

—Dharmesh (@dharmesh)