Something happening inside a 30,000 square foot facility in the heart of San Francisco’s financial district could one day be remembered alongside the moment Edison flipped the switch on the lightbulb for the first time.
Banyan Hill analyst Ian King believes a 247x “Adaptive AI” boom is about to be unleashed and one secret supplier stock is the best way to profit.
The Teaser
As soon as November 30th, Ian predicts Jeff Bezos, along with Elon Musk’s former right-hand man, will unveil a breakthrough that takes AI beyond the computer screen.
A former Salomon Brothers and Citibank trader with more than two decades of market experience, Ian King has been in the investment newsletter game since 2017 when he came to Banyan Hill Publishing.
We have reviewed some of his previous teasers, including the “x402 Project” Agent Economy Stocks and ASI Stocks, among others.
Ian calls this new form of artificial intelligence “Adaptive AI” because, unlike normal AI that can only answer questions, generate images, and write code, “this breakthrough will allow AI to learn, adapt, and operate with the physical world around it.”
Morgan Stanley estimates “Adaptive AI” will take a market that’s sitting at $101 billion today and “launch it $25 trillion higher over the long term.”
It sounds like Ian and Morgan Stanley are talking about some sort of operating system for robotics.
Our suspicions are confirmed moments later when Ian points to an Amazon Schedule 14A SEC filing referencing a “Project Prometheus.”
The reference is only a single paragraph in Jeff Bezos’ director biography:
Co-founded AI company Project Prometheus in February 2025 and has served as co-CEO since November 2025
There’s hardly any info on it available online either, showing that they take the whole stealth startup thing seriously. All we know is that it’s an “Artificial intelligence engineer designed to help invent, design, and manufacture complex physical products.”
Prometheus is housed at 101 Mission Street in downtown San Francisco, which is the facility in the heart of San Francisco’s financial district Ian mentioned at the start of the teaser, and it’s already raised $18 billion at a $41 billion valuation. By this measure, it is the fastest-growing startup ever.
However, unless we’re close personal friends of Jeff’s, it’s impossible to buy a stake in his new company directly. But Ian says he’s discovered a little-known business, 800x smaller than Amazon, that’s absolutely critical to making “Adaptive AI” a reality.
The Pitch
It’s name is only revealed inside a new special report titled “Bezos’ Secret Supplier: The One Company That Holds The Key to the 21,000% Adaptive AI Revolution.”
To get our hands on it we’ll need a subscription to Banyan Hill’s Strategic Fortunes newsletter, where Ian publishes his top investment ideas every month.
A subscription costs $49 per quarter (normally $199) and comes with a 30-day money-back guarantee.
The “Adaptive AI” Tipping Point
In 1913 Henry Ford’s assembly line unlocked an entirely new economic era, instantly giving 122 million Americans the opportunity to own a car practically overnight.
Now, for the first time, Adaptive AI is giving machines the ability to learn, adapt, and perform productive work in the physical world with little to no human oversight.
To be sure, industrial robots have been around ever since the Unimate worked on die-cast parts at a General Motors factory in 1961.
But apart from repeatable mechanical movements, the so-called ‘sensory layer’ has yet to be solved…until now.
Moravec’s Paradox
In the late 80s, robotics expert and computer scientist Hans Moravec described the physical robotics challenge brilliantly:
It is comparatively easy to make computers exhibit adult-level performance on intelligence tests and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility
It’s where the term Moravec’s Paradox originates and what engineers have struggled with ever since.
Ian believes Bezos and his crack team at Project Prometheus, including Kyle Kosic, the engineer who helped build xAI’s supercomputer in 122 days flat, may have finally solved the paradox.
By connecting the power of AWS, a digital brain, with DeepFleet’s foundational AI models, and Project Prometheus’ physical sensory hardware, we could have the biggest robotics breakthrough in decades.
The tipping point could come as soon as November 30th, which is the date of Amazon AWS’ annual re:Invent conference, and where the company has historically announced some it’s most disruptive technology breakthroughs.
Ian says it’s “not a long shot to think Project Prometheus could single-handedly double Amazon’s stock price in just 3-4 years.” But headquartered just 46 miles away from its headquarters in San Francisco is another company that could prove to be the linchpin of “Adaptive AI.”
Let’s find out what it is.
Revealing Ian King’s Secret Supplier Adaptive AI Stock
Besides geography, there are a few other things we learned about Ian’s “Adaptive AI” stock.
- This company builds specialized chips that “give machines their eyes.”
- Amazon hand-picked them as an official hardware partner for its computer vision platform.
- Currently, 800 times smaller than Amazon.
The company that matches all three criteria is Ambarella Inc. (Nasdaq: AMBA).
- Ambarella specializes in computer-vision AI System-on-Chips (SoCs), which are designed to give autonomous vehicles, smart cameras, and robots eyes.
- It is an official AWS Panorama Device SDK harware partner.
- At a market cap of just under $3 billion, Ambarella is over 800x smaller than Amazon’s $2.8 trillion market value.
Make 247x Your Money?
To make money in the market, you have to get at least one thing right.
To make a LOT of money, the kind Ian is teasing, many things have to go right.
For Ian’s “Adaptive AI” thesis to come to fruition, Jeff Bezos’ Project Prometheus stealth startup first needs to solve robotic perception and mobility. It then needs to partner with Amazon, which is less of a stretch, and finally, Ambarella, an Amazon partner, needs to also become a Project Prometheus supplier.
If all this happens, then we could potentially be looking at a mega multibagger.
The likelihood of just the first step working out is a wager that we simply do not have enough information to make at this time, as Project Prometheus hasn’t even announced what exactly it is working on.
This unravels the rest of the setup, but if we look at Ambarella as a standalone pick, without the Project Prometheus catalyst, we would be paying 7x revenue for a business that is expected to breakeven sometime during fiscal year 2027.
It’s balance sheet is surprisingly clean, with no debt, and only $21.5 million in long-term lease and other liabilities, against $272 million in cash. Annual operating expenses plus capital expenditures are around $63 million, so it should be able to comfortably reach profitability without any additional shareholder dilution.
Ambarella’s big bet is edge AI semiconductors or AI running directly on local devices like smartphones and tablets to enable real-time data processing, which is the way most AI agents will run in the future.
If Ambarella is able to keep significant market share in the system-on-a-chip (SoC) semiconductor market, today’s price could end up being a grand bargain. An asymmetric play for sure.
Quick Recap & Conclusion
- Banyan Hill analyst Ian King believes a 247x “Adaptive AI” boom is about to be unleashed and one secret supplier stock is the best way to profit.
- By “Adaptive AI“, Ian simply means advanced physical AI, which he believes Jeff Bezos’ stealth startup Project Prometheus could crack.
- The way to play it is by buying into a little-known stock whose name is only revealed in a new report called “Bezos’ Secret Supplier: The One Company That Holds The Key to the 21,000% Adaptive AI Revolution.” A subscription to Banyan Hill’s Strategic Fortunes newsletter is required to get it and it costs $49 per quarter (normally $199).
- Fortunately, we got enough hints and references to reveal it for free! It’s Ambarella Inc. (Nasdaq: AMBA).
- AMBA has a clean balance sheet, is projected to be profitable in the current fiscal year, and has a foothold in the growing edge AI semiconductor market. It’s a moderate risk, high-reward play.
How soon will AI processing move on-device? Leave your best guess in the comments.