My Claude-Code Revelation Moment: How the 2026 update turned a long-time AI skeptic into a believer. And why a major shift in the economy and jobs could now be a lot closer than you think

So far, the AI boom has been an expensive disappointment, involving massive hype, even larger debt and an expensive product that has done little to help businesses. But the latest release of Claude Code delivered a stunning and dramatic reversal in a crucial area for future AI growth. It also signals the death of one profession and is giving birth to another. And if AI continues to improve at this pace, the effect on the economy and jobs will be enormous — and will arrive sooner than many expect.

(Usual disclaimer: I'm just an investor expressing my personal opinion and am not an attorney, accountant nor your financial advisor. Consult your own financial professionals before making any financial decisions. Code of Ethics: To remove conflicts of interest that are rife on other sites, I/we do not accept ANY money from outside sponsors or platforms for ANYTHING. This includes but is not limited to: no money for postings, nor reviews, nor advertising, nor affiliate leads etc. Nor do I/we negotiate special terms for ourselves in the club above what we negotiate for the benefit of members. Info may contains errors so use at your own risk. See Code of Ethics for more info.)

I've been a skeptic of AI for a long time. And there are many good reasons to feel that AI financing is currently in a precarious, bubble-like state.

AI spends like a sailor on leave…

The dollar amount of investment pouring into artificial intelligence over the past several years is staggering. Estimates say that AI datacenter spending alone could reach roughly $400–$450 billion annually. And total AI infrastructure investment by major tech firms is expected to approach $650 billion in 2026. 

That's an almost unbelievable figure — equal to roughly 1% - 2% of U.S. GDP. And very few industries in history have successfully absorbed that much capital so quickly.

…while earning money like a pauper.

At the same time, the companies leading this spending are not just unprofitable -- they are burning through cash at levels that are unprecedented in all of corporate history. OpenAI (one of the most prominent AI firms), alone, is expecting massive losses from 2023 through 2028. They project losing $14 billion in 2026 and for growing to a total of $44 billion more... before hoping for a profit in 2029. And that burn rate alone dwarfs every existing record and easily qualifies as the most massive in history:

Anthropic (another prominent AI firm) is also losing billions every year (and as of last November is aiming for profitability by 2028).

On top of this, startups often take longer to reach profitability than their early projections suggest. And many technologies have followed a pattern where optimistic forecasts failed to materialize.

Circle of trust? Or circular firing squad?

Even more concerning is how the current AI boom is being financed. A growing number of deals involve complex circular financial relationships between the same small group of companies. These arrangements often involve cloud providers, chip manufacturers, and AI startups investing in each other while simultaneously serving as each other’s largest customers.

Bloomberg described the phenomenon in its March 11, 2026 report on “Circular Deals in the AI Boom”:

“The result is an increasingly interconnected web of dependencies between technology manufacturers and AI startups. The risk with these ‘circular’ deals is that they can create skewed incentives that may lead to bad decision making and magnify losses if demand for AI fails to match today’s lofty expectations. The stakes are high as the AI boom has sucked in gargantuan sums of money from debt and equity markets and buoyed multiple industries.” (bolding added)

Et Tu, Stock Market?

This also raises serious questions about whether today’s unusually generous stock market valuations are also out of sync with reality.

Since 2020, a shocking percentage of all the wealth created in the S&P 500 has come from just seven companies — the so-called “Magnificent 7,” all of which are deeply tied to the AI boom.

For example, NVIDIA has catapulted to the top of the size chart by selling the computer chips that power the major AI models. And those models are deeply connected to the others, with Alphabet owning Gemini and Facebook owning Lima. Microsoft is a ~27% investor in Open AI. And Amazon is a ~20% investor in Anthropic, etc.

And periods of extremely narrow market leadership like this are historically unusual. Similar episodes occurred during the “Nifty Fifty” era of the early 1970s and the internet boom of the late 1990s. And in both cases, those periods were eventually followed by market crashes of roughly 50%.

That doesn’t mean a crash is inevitable. But it does mean the market has become unusually dependent on a very small number of companies. And that kind of concentration is rarely healthy.

FOMO (“Fear of Missing Out”) leads to billion-dollar corporate AI disappointment

Worse has been what AI has delivered so far.

Many companies live in fear of falling behind. So the average large company invested a whopping $110 million in AI last year. And together they spent billions in 2025. So what do companies have to show for this unprecedented spending spree? Studies have consistently found the answer is “very little”: 

1. In October 2025, Boston Consulting Group reported that 95% of companies had derived no meaningful value from AI initiatives.
   

2. Around the same time, a Gartner survey found that 72% of firms had achieved no positive return on investment from their AI deployments.
   

3. And in February 2026, a National Bureau of Economic Research survey of roughly 6,000 executives found that more than 80% of companies reported no productivity gains from AI despite billions of dollars in spending.

“Tell me about it”

And many of us have had similar, disappointing personal experiences with AI.

Anyone who has used AI extensively has likely seen it confidently deliver answers that are completely wrong. They've seen it fail at tasks as basic as elementary school math. And they've seen it frequently produce explanations that sound convincing but collapse under closer scrutiny.

These experiences became so common that this humorous meme has been very popular on social media:

This perfectly expresses the frustration that so many of us have felt when interacting with AI tools.

"Keep that AI away from my code"

Before my second career as an entrepreneur, I was a professional software developer. and I spent decades honing my craft and ultimately advanced into senior developer and project leadership roles. Later, when I started my own companies, I often served as the IT manager as well.

About six months ago, I decided to test the AI hype myself. And I used it to build a Monte Carlo simulation program to analyze our personal finances.

The experience was excruciatingly frustrating.

On one hand, the AI occasionally did things that were genuinely impressive. For example, it could refactor sections of code — something that is error-prone for human developers and can take weeks — in just a few seconds.

But most of the time, it felt like working with a genius who had the memory of a goldfish and the attention to detail of a slacker.

It constantly forgot what it had just done a few minutes earlier. I had to repeatedly re-upload the same files hundreds of times. It would forget important instructions and introduce unnecessary mistakes that broke working code.

It struggled to maintain architectural consistency. It ignored established protocols. And once the codebase became moderately large, it had major difficulty extending or debugging the program without creating new problems.

From the perspective of a professional programmer, everything it produced was completely unacceptable.

And after a few months of struggling with workarounds after workarounds, I finally gave up.

So once again, AI had fallen far short of the enormous hype surrounding it.

Writing on the wall at the end of 2025

So as recently as December 2025, my view was that AI would likely follow the same path as many other over-hyped technologies that ran into unexpected limitations.

History is full of many famous examples:

1. In 2016, Elon Musk said Tesla would demonstrate full autonomy self driving cars with a coast-to-coast drive:

   “We'll be able to do a demonstration drive of full autonomy all the way from LA to New York… dropping you off in Times Square… and then have the car go and park itself by the end of next year.” (i.e., by 2017) (bolding added)

   That demonstration never occurred. And nearly a decade later, fully autonomous consumer vehicles still don’t exist.
   

2. Since the 1930’s, nuclear fusion has famously been described as perpetually out of reach, “just a few decades away.” As one widely repeated saying among physicists puts it: “Fusion is the energy of the future — and always will be.” Even today, commercial fusion power plants remain years or decades from reality.
   

And by late 2025, I felt it was likely that the AI boom would end in a similar way.  And I thought an ugly stock market crash was inevitable.

Revelation Moment: 2026 Claude Code changes everything

About a month ago, all of that changed in a relative instant. An acquaintance told me that Claude Code had “gotten a lot better.” So I tried it… yet again ... and mentally prepared for more frustrating disappointment.

But this time, the results were so surprising, that they rocked my world.

Monte Carlo redux..

I gave Claude Code my partially completed Monte Carlo program — the one that the previous version of AI had struggled with for months. And I asked it to finish it.

To my great surprise, it fully completed it within a few hours.

Even more surprising, the resulting code looked very clean. No messy spaghetti code -- and very different than the monstrosity it had accumulated during earlier attempts. The architecture was consistent. And the program worked.

Also, Claude Code didn’t repeatedly forget the files or instructions I had given it. The constant “goldfish memory” problem had (largely) disappeared.

And this time, the end-result were actually good and usable. That was unexpected. So I thought "What else could I try"?

And I decided to give it something a lot more difficult.

Chasing planes

Back in 2024, my wife and I were looking at potentially buying a lot to build on. And we treasure quiet, so I hired a developer to build a custom program to track airplane flights (to identify good and bad areas of town to target). It was a fairly complex project because it involved tying into multiple APIs from different companies: Google maps, airplane tracking data, etc. I found and hired a great programmer online and he finished it in just six weeks. It cost me only $3,500, which was an absolute bargain and only because he lived overseas. A domestic coder would have cost me a lot more. So in the end, I was very happy with the developer, the quality of the work and the price.

Now I wondered, what would Claude Code do with this much bigger challenge?

To make the comparison fair, I gave Claude the exact same requirements I sent my human programmer. It first asked me some clarifying questions. This was a nice improvement from the earlier version, that would shoot from the hip and waste time going in the wrong direction. So I answered them... and then it started working.

Amazingly, Claude created the entire program (and more) in about five hours. And it cost roughly $0.69 (based on 5 hours cost out of my $100 monthly Claude Code subscription).

And more importantly, the quality was remarkable.

1. Claude required far less detailed instruction than the human developer had needed. It quickly figured out complex API integrations and even identified an error in the API documentation that I myself had not noticed.

2. It also produced a more visually appealing and professionally polished user interface than the original version.

3. And it added several useful features that I hadn't thought to request (and never would have requested from a human developer because of the cost).

And it also was more robust to errors and worked great. Here's the map view it created (with a lot of sophisticated features that I didn't even think of myself, but appreciated):

Map view

...and the flight view (with a very clean and easy to view interface):

Flight view

So this completely blew me away.

And what's more...I could clearly see a fairly easy path to exponential improvements on TOP of this. The main reason it took 5 hours is because Claude couldn’t see my desktop (where I ran the code).  So I spent that time as an intermediary between the AI and my computer — running the code, capturing errors, and feeding them back to Claude.

Once AI tools can directly view and interact with a user’s desktop — (and this is something already being worked on in products like OpenClaw, which we'll discuss later on) — the debugging process will become automated as well. And AI being able to self-iterate will be yet another game changer. Had I been able to iterate, the project could have been cranked out in just 15-30 minutes. And that would have cost closer to $.03 to $.07. That's essentially "free".

[Update March 24, 2026]: Claude Code announced a breakthrough beta product called "preview servers" several days ago that purports to allow Claude Code desktop (also in beta) to self-iterate. I've working with this in depth the last several days. And I can report that it does indeed allow self-iteration on certain types of programs -- like web apps and certain desktop apps with very limited GUIs. For example, Claude Code can see the website, click buttons to test it, take a screenshot of the error, and then try again to fix the bug -- all without human intervention. This is an enormous leap forward versus what was available just a month ago. And this reinforces my belief that improvements are currently coming at an unprecedented and accelerated speed. And I have no doubt self-iteration will soon come to many more types of software development.

Death of a Profession...

The whole experience was astounding and stunning.

Programming is something I spent decades learning and perfecting. It took me 4 years of painstaking work to get a degree. And then it took years of experience in industry to develop the skills and habits to design reliable software systems. And it allowed me to make a good living for many years. And it was difficult work and so I also took pride in it (and also the professional awards and certifications I accumulated while doing it).

Yet, here was a machine performing the same work — but doing it exponentially faster and at virtually zero cost!

It wasn't perfect. But in vast majority of areas, it ranged from a lot better to exponentially better than a very good human programmer. And this skill will only improve.

So on one hand, I felt shocked and sad. I wouldn't want to be a computer science graduate in 2026. And I feel the way we teach programming -- and many other degrees -- will need to be completely revamped for the new reality. (Although that's a whole other topic and for a later article). But at the same time ,I was also filled with a feeling of wonder and excitement at the unprecedented and enormous new world of possibilities that this opens up -- for companies of all sizes and for the creation of brand-new professions.

The Coming Software Explosion

Since the creation of the computer in the 1800's, software development has always been tedious, slow and expensive. And the needs of people and companies for it has been ravenous. As a result, the industry has perpetually fallen woefully short of fulfilling all the enormous demand for creating programs. AI is now uncorking a two-century-old bottleneck and turning the old dynamic on it's head. And this will unshackle all the pent-up demand for software creation. In my opinion, we are about to see a software explosion the likes of which we've never seen before. And this will enable the creation of an unprecedented boom of young, dynamic companies -- and brand new professions too.

Rise of the entrepreneur and the micro company

Already, many entrepreneurs and entrepreneurs-in-waiting are expressing giddy excitement at the new-found abilities that are suddenly in their grasp. Entrepreneur Mark Sear said on Linked In:

If you've ever thought "I could build better software for this", the barriers that stopped you are gone. The tools are here and improving weekly. .. A single person with domain expertise and tools like Claude Code can build what needed a 20-person team three years ago... ... Visionaries without funding can now build their dreams. ... The next great software companies won't be born in Sand Hill Road pitch meetings. They'll be built by domain experts who finally have the firepower to match their vision.

Famed startup incubator Y-combinator acknowledged the tectonic shift in creating new companies a few weeks ago, when they said:

The way startups are built has shifted quickly... AI-native companies can now be built faster, cheaper, and with more ambition than ever.

And a few days ago, the Wall Street Journal said:

Creating apps and websites used to be the rarefied world of tech developers. No more. During the tech boom of the 2010s, coding teams here spent months building apps that changed the way we live. Now, that’s just a Saturday night for an 11-year-old.

In my opinion, AI's ability to code will democratize access to software and allow the creation of more new companies than ever before (and bring an increase in creativity and dynamism to the economy). And the scope of this change goes far beyond just entrepreneurs and startups. When software development is virtually free and very fast, it will also unlock and enable new jobs that never existed before.

The birth of a new profession: The AI-enabled "Builder"

That same Wall Street Journal article gave a recent example of the experience of Clayton-Stankowski.

Clayton-Stankowski was until recently a project engineer at a Bay Area construction company. Surrounded by friends who work in tech—many using automated coding programs—he started thinking about what he could automate at work. He also wondered why he was paying $10 a month for a habit-tracking app when he could build one himself. So he did. He was hooked. Soon he was identifying ways to save time and money using AI at the office. He even pitched a new role for himself: AI engineer. He started last month.

To put that into perspective: Everything Clayton-Stankowski did would have been impossible for him to do only 6 months ago. Now, he's solving business problems that couldn't be tackled before and has a job that didn't exist before.

And virtually every company (both large and small) is in a never-ending journey to remain competitive, improve efficiency, and enter new markets. And virtually all have a similarly ever-growing list of tasks that they need done. AI coding now makes solving those problems possible.

So I believe we are seeing the rise of a new profession: using personal domain knowledge and creativity to solve business problems with the use of AI tools.

Clayton-Stankowski used the word  "AI Engineer" to describe it. But that's a misnomer because that job already exists and means something else (a software programmer who develops the AI models themselves). His job might more accurately be called an "AI-Enabled Software Builder." In a recent Business Insider article, Windsurf CEO Varun Mohan said that computer programming in general is evolving into the profession of the AI-enabled "builder":

"This notion of just a [software] developer is probably going to broaden out to what's called a builder. Traditional developers may no longer be the only ones building software, he said: "I think everyone is going to be a builder."

Builders, Builders Everywhere and Not A Coder To Be Found ...

And this is just the beginning. Later we'll discuss how software is merely one of the first fields to get disrupted by AI -- and we'll see which ones are likely to follow next. For now, the important point is that AI today is currently in the "red" area. And the "blue" is where AI has the theoretical potential to spread to in the future (which I've highlighted with a yellow outline):

So if AI tools improve and move into those blue areas... the need for Builders will expand into them, too.

San Francisco is the birthplace of the majority of new startups and of many business trends that eventually spread across the country. And in an article about the city's trends, this Wall Street Journal article describes how, almost overnight:

Now everyone’s a builder: marketers, product managers… tweens

Mateusz Litarowicz is a marketing and product design professional. And he published an article a few week ago saying companies are already hiring for this:

Will AI make us all product builders? Is the market really moving in this direction? In my opinion, absolutely yes. ... this shift is already happening. ...Some organizations are already hiring for roles like this, while in others, it may remain out of reach for the next few years due to their size, risks, product complexity, or simply a lack of awareness.

To be fair ... it's far from guaranteed that AI will fulfill its promise and spread to all the fields (marked in blue in the chart above).

So the future of builders across industries isn't guaranteed either. And we'll talk more about those possibilities, later. But what AI has already achieved has revolutionized the programming field profoundly.

The Economic Theory of Close to Zero Marginal Cost

The emergence of the software builder is a brand new thing — but it’s really just an illustration of an old idea.

When the marginal cost of something approaches zero, usage explodes and more people start producing it.

We’ve seen this before. The internet drove the cost of publishing toward zero — and suddenly anyone could create content. That reduced but didn’t eliminate the traditional creators. It also created many new jobs for a new generation of creators: influencers, bloggers, YouTubers, Twitch streamers, internet journalists and more. And it also created new methods of distributing content, such as streaming on Netflix, Prime, etc., and posting to social media accounts on Facebook, Twitter/X, etc. (with new jobs in all of these areas as well).

Now AI is doing the same thing to software.

As the cost of building software collapses, the result isn’t fewer creators — it’s more people building, just at a higher level of abstraction.

“This isn’t your daddy’s AI”

So that's why I’ve gone from being an AI skeptic to an AI believer.

And I now believe AI is likely to become one of the most economically consequential technologies of the next several years.

And if you looked at AI as recently as six months ago and felt it was awful – I highly recommend that you take the time to download the latest models and see what's been happening more recently.  And just so you know: the free versions (like you see on Google search and in Windows tools) are almost all older versions that are way behind the state of the art. If you want to understand what's happening now, you have to pay for a subscription.

Also, I've noticed that different models can behave very differently from others. For example, I've found Claude Code to be particularly strong at programming tasks, but awful for research. And ChatGPT performs better than Claude with research, but its coding isn't so great. So it's a mistake to try just one model and conclude "AI can't do that." Another model just might already be doing it.

For people who want to follow the industry itself closely, I also recommend subscribing to “The Information,” a Silicon Valley news service. The subscription is expensive — roughly $749 per year. But their reporting on the AI industry is consistently deeper and earlier than most other publications.

What will be the consequences of all of this?

In Part 2, I’ll talk about what this means for technological breath-throughs. And we'll look at both the case for rapid change and the significant obstacles that could stall or stop AI's potential from being realized. And we'll look further at what would happen if AI does succeed.

Click here for Part 2. ------------------------------------------------------

Discussing AI and more

Private Investor Club members have been discussing AI in depth.  And there's also been detailed discussion on thousands of sponsors and deals (including due diligence and real-life investor experiences).

• If you're already a member, click here to discuss further:

• If you're not yet a member, then joining is free.

  • To protect all members and keep the conversations useful and confidential, all applicants are required to verify that they're solely investors (and not investment sponsors, platforms or their affiliates)

  • Click here to learn more or to join.