2026 is about the get wild.

I have called Asymmetric Investing my investing diary.

That includes not only what I’m doing with my money and why, but also the way I work through investing concepts and data.

Today, I want to take you through some of the things I’m thinking about AI and energy. This isn’t a definitive answer to what will happen with either. Instead, it’s the thoughts and questions I have about the future.

I hope you find this useful in understanding how I’m thinking about the future and where I’m excited and worried about technology and energy.

If you’re getting value from posts like this and want to get access to my market-beating Asymmetric Portfolio, become a premium subscriber below. Thank you for your support!

This week, I built a tool to demonstrate how quickly wealth can compound depending on your starting balance, monthly contributions, and time horizon at different growth rates.

The Asymmetric Portfolio, for example, started from nothing, and I invest $500 per month. If I generate a 15.0% annualized return, the portfolio would be worth $3.46 million in 30 years.

After two years of returns closer to 30%…I hope I can beat a 15% return. But historically, that’s a high bar.

You can see an image of the tool below and play with it yourself by clicking on the image. I don’t show this to show off my coding skills. I have none. And if you look closely, the numbers below are correct, but the graph is wrong, so AI is still making mistakes.

I point this out because building this tool required exactly zero coding skills. I literally put this prompt 👇 into Claude and then asked it to make that output into HTML. If this doesn’t show the commodification of a huge swath of knowledge work, I don’t know what does.

I’ve been writing for over a year about my skepticism about investing in AI. But that’s about the returns on investment in data centers and what business models will work. I’ve never questioned that AI will impact the world.

AI is likely to be just as impactful as the internet was starting in the mid-1990s.

What I’m starting to see more clearly is where there’s going to be returns for AI usage and disruption in many jobs and industries.

We’re seeing is a paradigm shift in the return on agency.

If you want to build something…you can.

You can have 100 AI agents working for you. You can have access to skills that were once incredibly expensive.

Not everyone has that kind of agency, but for those that do the sky is the limit. That’s the opportunity and the threat.

Increased agency will impact the operations and incentives of corporations. We’re already seeing job cuts at big tech companies because the best engineers can be more productive writing code with AI.

But AI isn’t just a negative for jobs. It will unlock opportunities for people with the most agency.

I’ll use myself as a very simple example.

I built a YouTube channel that now has over 45,000 subscribers. I record content, edit content, make thumbnails, negotiate sponsorship deals, etc. On top of that, I write this newsletter, I run the website, I designed (with some help) how the website works, I run payments, I answer reader questions, etc.

I do it all!

Not because I’m a genius.

Not because I work harder than anyone else.

I simply have high agency and a motivation to be independent. And in 2026, there are higher returns to my agency than I (or anyone else) could have had a decade ago.

I didn’t need to hire a writing editor (I have Grammarly).

I don’t need a video editor (I have Descript).

I don’t need a software developer (I use Beehiiv and vidIQ).

I don’t need a social media team (I have Hypefurry).

I’m not building a business with agents the way tech bros talk about them, but the platforms and tools that have been built provide me with incredible leverage on the work I do. Everything I can’t/don’t want to do is the click of a button. 10-20 years ago, I would have had to hire someone to do a lot of this work.

Below is a look at the video editor I use, and it’s literally a button click to cut out dozens of word gaps, retakes, and filler words from a video.

Another click and I have chapters for YouTube.

Another click and the video is published to YouTube.

That’s hours and hours of time saved. This kind of productivity improvement isn’t new in technology. What’s new is the scale of the productivity improvements and the breadth of what’s available without having a big team.

I don’t know anything about programming, but I can create a computer program and put it on the website.

I don’t know anything about graphic design, but I can create great images.

Some people call these agents. I think of them more like skills. Skills I can offload to an AI and not ever have to learn myself.

What’s the implication of these agents/skills? In the corporate world, the short-term impact is clear.

Meta has been laying off people.

Amazon is laying off people.

Block is slashing staff.

Is AI a job killer?

Kind of…

For big companies, it’s going to be a way to squeeze more productivity out of people. At least, that’s the goal. As we saw with the PC and internet, getting from here to there will be harder than a lot of companies think.

More importantly, I think AI is an agency enhancer. That’s good for the individual, but may not be as positive for the average corporation.

What’s the incentive to use an agency enhancer in Corporate America? So, Mark Zuckerberg can make a few more points of free cash flow margin?

There’s a tension between productivity and hyper agency-enhancing tools in large businesses.

Where there are no limits to enhancing agency is in small businesses — especially one-person businesses — where the incentives are aligned.

If I put out an extra video or test an ad or post more on social media, the value accrues to me.

Unlike writing or video production in the past, the agency of a person has a direct reward.

Does that same incentive structure exist everywhere?

Probably not.

So, what’s next?

I think we see the Smiling Curve on steroids.

A high agency person may be able to go from making $200,000 in a corporate job today to making $20 million in five years. Companies may be forced to pay up for good talented people that get stuff done.

As they should!

But what about the person who just wants to go to work, be productive, and go home? People in the middle of the curve may be washed out, even if they’ve “done everything right.”

What happens to people who have gone to school, climbed the ladder, and built expertise in areas that were once valuable, but are now being commoditized by AI?

In this case, the architect doesn’t become the carpenter. The carpenter becomes the architect. The carpenter is the one who — with enough agency — can increase their value exponentially.

What happens to the architect and the designer? They’re the ones who have to adapt…

Another big question surrounds how corporations respond. Why are 10 people doing the work that 1 person could…if that person has enough agency?

But if you’re that 1 person…why work for a corporation, unless they give you 10x the money?

Will Best Buy pay a great programmer or a high agency marketer $2 million a year because they are that valuable?

I don’t have the answers, but this is what I think about when I see layoffs and employment data. Workers with high agency are going to be fine, and there will be some amazing businesses built with relatively few employees and a large number of AI tools and “agents.”

But who drives the economy from here to whatever the future looks like when the architects, designers, lawyers, and doctors are out of jobs? Will the high agency carpenter, the plumber, and the motivated YouTuber be enough to build a new economy?

Maybe…

But it’s not likely to be an easy path from here to there.

If oil isn’t on your mind right now, it should be. We may be in for one of those generational moments in the energy industry as Iran and the rest of the Middle East sees both supply and transportation impacted.

Here are some things I think I know that may be wrong.

If you haven’t spent time playing on the EIA website…maybe you won’t find this fun. But I think it’s fascinating.

This is a table I’ve had bookmarked for more than a decade, and it tells a lot about what’s going on in the world of oil.

For this portion, I want to focus on net imports. You can see that in 2005, the U.S. was a net importer of 12.5 million barrels of oil per day. Today, we’re a net exporter of 3.1 million barrels per day.

That sounds great, but you can also see that the U.S. still imports about 8 million barrels of oil per day. This is because we’ll import oil, refine it, and then export the refined product to other countries. There’s a global mismatch on where refiners and demand are located, and also what type of oil refiners can actually refine.

I will also note that imports from the Persian Gulf have gone to almost zero, and OPEC overall is about 5% of U.S. supply.

There’s just one problem, in the U.S. at least.

Oil is fungible (for the most part).

That means a barrel of oil in the U.S. is equivalent to a barrel of oil in China, or anywhere else.

Demand is relatively inelastic (demand doesn’t change much with price), so prices can spike with even a 1% or 2% change in supply. A 20% hit to the global oil supply would be devestating.

There are band-aids like releasing oil from the petroleum reserve, but that’s going to fix the problem for a few weeks, not long-term.

This is different from natural gas, which is more difficult to move from one country to another. Pipelines exist, but natural gas isn’t dense, so to move it from the Middle East to Europe, for example, it needs to be condensed into a liquid, which is costly.

That’s why you’ll see the impact of the Iran war at the gas pump, but not in your heating bill.

We’ve already heard reports that 17% of Qatar’s LNG facilities were hit and it could take years to bring capacity back online.

20% of global oil supply travels through the Straight of Hormuz and a large percentage of that is going nowhere.

Some of Iran’s oil production seemed to have been impacted, although we don’t know how much. The country produces over 3% of the world’s oil, so taking them offline could result in a huge jump in prices.

But the problem doesn’t end with the disruptions today. The problem is the disruptions long-term.

Qatar said it may take 5 years to bring its supply back online.

If Iran’s oil production is impacted that could be years of reduced supply.

Then there’s disruptions in Saudi Arabia and the UAE.

The U.S. can increase domestic production, but that’s months to years to get that going.

Offshore fields are also plentiful, but those take multiple years to begin producing oil.

The reason to keep an eye on the the war is that millions of barrels of oil could be taken offline for years if things get worse. Saudi Arabia, Iran, Iraq, the UAE, and Qatar account for over 30% of global oil production, so the stakes are high.

Supply changes could last years, but when prices go up, demand can changes can be permanent.

Look at the table above and notice how much oil the U.S. consumed in 2005 and how much it consumes today.

Yes, the U.S. consumes less oil today than it did in 2005. That’s because of:

Notice that changes in usage followed the Iranian oil shock in 1973 and the oil price spike from 2001 to 2008.

After both price shocks, demand changed forever.

Finally, I’m thinking about what the 2nd and 3rd order affects of higher oil prices are on the economy. If people are spending double what they were a year ago at the pump:

High oil prices are like a tax on the consumer. It makes everything more expensive and leaves less money for people to spend.

And if this war impacts oil prices for the next few years, a recession isn’t out of the question. Companies are already happy to use AI as an excuse to cut headcount and if the headcount that’s left working has less money to spend on stuff…the economy suffers.

I’m not predicting an economic collapse because of Iran, but I want to be aware of the risks ahead. I’ve lived through a stock market that was flat throughout the 2000s, impacted by both the Iraq war and rising oil prices that (I think) broke the economy and exposed the financial crisis in 2008.

I’m old enough to remember the long lasting impact of a stagnant economy and market following the 1973 oil crisis, which ultimately led to inflation and skyrocketing interest rates in the early 1980s.

Things can go south very quickly. If you began investing in the last 17 years, you’ve seen a period of unprecidented calm. That may not last forever.

These are all things I’m thinking about. I’m hoping for the best, but we live in a reality that may get worse.