One overlooked corner of the AI Boom is still wide open – and he’s found a young company already putting the technology to work.
Listen to the audio version of this article (generated by AI).
Last weekend was filled with calls from prominent AI leaders to pump the brakes on frontier development. Understandably, Wall Street began worrying about what a slowdown could mean for the hundreds of billions pouring into AI infrastructure.
But in today’s Friday Digest takeover, our technology expert, Luke Lango, makes an important distinction. Building tomorrow’s AI is only one side of the boom. The other is figuring out profitable ways to use the extraordinarily powerful AI we already have – and that process is still in its infancy.
Luke believes that could create an entirely new crop of winners as entrepreneurs apply existing AI to a variety of industries. He highlights a fascinating example – a young private robotics company that’s already putting AI to work in the real world.
Luke recently recommended the company to members of his Venture Capital Investor service, and he walked through the opportunity – and its risks – in his free 2026 AI Megadeal Event. You can watch the replay right here.
Bottom line: Even if the race for smarter AI is hitting a speed bump, the race to make money with today’s AI is just getting started.
I’ll let Luke take it from here.
Have a good evening,
Jeff Remsburg
Hello, Reader.
The AI industry just had one hell of a weekend.
It started last week when 27-year-old Anthropic researcher Jacob Coxon quit and accused Anthropic and OpenAI of racing toward superintelligence while “gambling with our lives.” Then on Saturday, Anthropic CEO Dario Amodei published an essay called “We Must Pace the Frontier,” arguing that the industry needs to slow the development of increasingly powerful AI models.
OpenAI CEO Sam Altman agreed. So did Elon Musk.
Wall Street responded pretty much as you’d expect. AI stocks sold off as investors started asking what a deliberate slowdown could mean for the hundreds of billions of dollars pouring into chips, data centers, power plants, and everything else supporting the AI buildout.
I’ve spent a lot of time thinking about that question. And while I don’t think this changes the direction of the AI Boom, it could change the speed.
I’ve said for months that politics, regulation, and public concerns about AI safety could put some speed bumps in front of the industry. We may be seeing the beginning of that now. Some of the more aggressive forecasts for how quickly AI develops may have to come down.
But there’s another part of this story I think investors should understand.
Amodei is primarily talking about slowing the development of frontier AI – increasingly powerful models capable of reasoning, coding, operating autonomously, and even helping researchers build better AI. That could mean longer development cycles, more safety testing, new rules, or limits on some types of training.
Meanwhile, companies all over the world are still figuring out what to do with the AI we’ve already built.
And there’s a lot left to figure out. For investors, I think there’s a lot of money left to be made there, too.
Because the next big AI winner doesn’t necessarily have to build a smarter model than OpenAI or Anthropic. It could take the extraordinary AI we already have and find valuable new ways to use it.
Today, I want to show you why I think that opportunity could keep growing even if frontier AI slows down. Then I’ll tell you about one young private company I recently recommended – a food-service robotics startup that’s already putting AI to work in the real world.
We’ve Barely Started Putting AI to Work
Think about the AI models available today. Businesses are already using them to write software, review documents, answer customer questions, analyze medical images, design products, and automate parts of their operations. Most companies are still early in that process.
AI requires computing power for two main jobs. Training is how developers build and improve a model. Inference is what happens every time somebody puts that model to work. Deloitte’s 2026 outlook projected that inference could account for roughly two-thirds of AI computing this year, up from about half in 2025.
So even if tomorrow’s AI takes longer to arrive, more people using today’s AI can keep demand growing for servers, memory chips, networking equipment, cooling, and electricity.
That’s one reason I remain bullish on AI infrastructure stocks.
But I’m also interested in the companies doing the actual using.
There are millions of businesses out there applying AI to real problems.
One of the companies I’ve been studying recently is doing that with robots.
And food.
Teaching Robots to Learn
The company I mentioned earlier started in food-service robotics. Its robots are already working in real commercial locations, serving actual customers.
I’ve visited one of those locations myself. I watched a robot server take an order, prepare it, and deliver the finished product. And I came away impressed.
But the robot server itself is only part of what interested me. Behind that business, the company has spent years developing what amounts to a training academy for robots.
Humans learn physical skills largely by watching other humans.Someone shows you how to do something, you try it yourself, they correct you, and you get better with practice. Robots have traditionally required specialized engineers to program their movements, which makes teaching them new physical tasks expensive and painfully slow.
This company is working on a different approach. Its AI system uses human demonstrations to teach robots new physical skills. The company says it can teach a robot some new hands-on tasks in as little as 30 minutes, without an engineer programming every movement.
That’s when this became a much more interesting company to me.
Food service gives these robots a place to learn and improve every day. But if this training technology works at scale, the same approach could eventually teach robots to handle products in warehouses, work with equipment in factories, perform tasks in healthcare, and take on other complicated physical jobs.
Now we’re talking about a much bigger potential market.
The Opportunity Beyond Smarter Models
This is the part of the AI Boom I think could get overlooked amid all the headlines about superintelligence, slowing down frontier development, and even the possibility that advanced AI could threaten humanity.
We already have extraordinarily capable AI, and businesses are putting it to work fast. Government data tends to produce more conservative adoption estimates, while business surveys have found anywhere from roughly 70% to nearly 90% of companies using AI in some fashion. The exact percentage depends heavily on what you count as “using AI.”
The larger point is that adoption has a long way to run. Entrepreneurs will spend years finding new applications for today’s technology, and successful ones will create demand for more computing power, infrastructure, robotics, and technologies we haven’t even thought of yet.
That’s why I’m paying close attention to young companies like the food-service robotics company I just described.
And increasingly, I’m looking for some of these companies while they’re still private. New technologies often start with small companies solving one narrow problem extremely well. If that technology proves valuable, a larger company may eventually decide it’s faster to acquire the business than spend years trying to re-create it. For the early investors who backed that young company, an acquisition can provide the payday long before an IPO ever arrives.
That’s one reason I’ve started looking beyond the stock market for AI opportunities. It gives me a chance to study promising young companies while they’re still building – and, in certain cases, invest alongside them.
Of course, investing that early comes with plenty of risk. The company I’ve been telling you about is young, it’s losing money, and its robot-training technology is still early. Its current valuation also puts a hefty price on growth that still has to materialize.
That’s where my PPT framework comes in.
Whenever I evaluate a young, privately held company like this, I don’t have years of SEC filings or a stock market history to look at. Instead, I start with three things: the People building it, the Product they’ve created, and the Timing of the opportunity. I call that my PPT framework.
This company checks some important boxes. Its CEO previously built a computer-vision startup that was acquired by Amazon.com Inc. (AMZN). Its robots are already operating in the real world. And its robot-training technology is arriving as major technology companies pour money into robotics and physical AI.
That’s why I recently recommended the company to members of my new Venture Capital Investor service.
During my free 2026 AI Megadeal Event, I walk you through the company from top to bottom. I go over its founders, food-service robotics business, robot-training technology, financials, and risks… and the reasons I decided to recommend it.
I also explain how individual investors can invest in private companies like it. If you’ve spent your investing life buying stocks through a brokerage account, this will probably be unfamiliar territory. I’ll show you how it works, what you’re actually buying, and what you should understand before putting your own money into one of these opportunities.
Watch the free replay of my 2026 AI Megadeal Event here.
Sincerely,
Luke Lango
Senior Investment Analyst, InvestorPlace
P.S. I’ve worked with Luke for a long time, and when he gets interested in a company, he likes to kick the tires himself. In this case, that meant visiting one of the company’s locations and watching its robots work firsthand. It’s a cool story, a fascinating young company, and a side of AI investing most of us rarely get to see. Check out Luke’s free event here.
