There is a common pattern in new technology cycles, and it goes like this:
The innovation itself appears. Then, a bottleneck emerges. Next, capital floods in to solve the problem. Finally, the regime changes.
We saw this “regime change,” or complete reorganization of stock market winners and losers, in the in the dot-com bust phase.
Capital rotated out of the high-profile names and into a variety of other sectors, including base metals, precious metals, energy insurance, and utilities. Those sectors delivered solid double-digit or triple-digit returns over the early part of the 2000s, even while the Amazons, Intels, and Ciscos of the world fell 80% or more.
Another regime change is happening now.
Since the early AI revolution, the Magnificent Seven companies have been sat securely on the throne. The group includes Alphabet Inc. (GOOGL), Amazon.com Inc. (GOOGL), Apple Inc. (AAPL), Meta Platforms Inc. (META), Microsoft Corp. (MSFT), Nvidia Corp. (NVDA) and Tesla Inc. (TSLA).
But their seat is soon to be usurped. We are starting to see a rotation out of some of the highest profile, high beta tech stocks and into more real-world, asset-backed sectors.
This dynamic will only increase as we move forward. That means it’s important to own the companies that are providers or suppliers to this massive AI buildout, rather than the companies that spending the money to do build.
And this regime change will soon shift into high gear.
That’s why, inside this report, I list three of my favorite stocks that are significantly outperforming the Mag 7 so far this year.
These “asset heavy” companies are a bellwether of the regime change that I predict will soon kick off after the next wave of earnings announcements from the AI hyperscalers.
Now, I want to be clear that I am not listing these companies as “Buy” recommendation today. Rather, I believe they will rise in rank as Big Tech household names continue to fall and, therefore, deserve our attention.
Let’s dive in…
Mag 7 Killer No. 1: A Prolific Energy Producer
Our first Mag 7 killer is one of America’s most prolific energy producers, and it’s firmly positioned to fuel America’s AI buildout.
As the big tech companies build ever-larger and ever-more-numerous data centers, they are struggling to secure the dedicated power supplies these centers require.
Because of power bottlenecks like these – both current and prospective – Big Tech is turning to every and any possible power source to satisfy their needs.
Even nuclear power is making a comeback. Amazon, Alphabet, Microsoft, and Meta have all inked deals to obtain dedicated nuclear-powered electricity.
But nuclear power alone won’t solve the problem. “Not in my back yard” issues, coupled with the lengthy permitting and construction process, ensure that nuclear power will not become more than a partial solution.
That’s why these big tech companies are also implementing an array of other power-generation technologies – ranging from solar-plus-storage to hydrogen fuel cells to natural gas “peeker plants.” Even geothermal systems are powering some data centers.
Natural gas is the clear winner in the push to generate power directly at data centers
According to a recent report by Cleanview, a market intelligence platform, nearly 75% of the power equipment planned to be used on site at data centers is natural gas.
The natural gas market offers three major advantages over competing technologies…
- It is abundant.
- It is cheap, especially in the Delaware Basin, one of the world’s most productive oil and gas regions located in West Texas and southeastern New Mexico.
- It is a proven technology with relatively rapid permitting processes.
Data center demand for natural gas could become especially acute in the Delaware Basin, with a new “Data Center Alley” potentially blossoming in the region.
That brings me to Devon Energy Corp. (DVN).
As one of the leading producers in the Delaware Basin, Devon is well-positioned to benefit from structurally improving pricing trends in the region.
For the last few years, the Delaware Basin has been rapidly boosting its production of both oil and gas. Unfortunately, gas volumes have overtaken pipeline capacity. As a result, much of the gas from the Delaware Basin is “stranded” — the oil and gas industry’s polite way of saying “worthless unless you can move it.”
Producers flared it. Trucked it. Discounted it to oblivion. As such, the Delaware Basin has behaved for years like a brilliant student stuck in detention. It held enormous potential, but had no way to express it.
But detention is ending.
Two major pipeline projects are improving the economics of the Delaware Basin, especially for Devon.
First came the Matterhorn Express, a 580-mile pipeline capable of moving 2.5 bcf/day from Waha to the Houston area. Devon secured long-term transport capacity from this pipeline, once it entered service earlier this year.
Next comes the Blackcomb Pipeline, a 365-mile system that will send gas south to the Agua Dulce hub near Corpus Christi – a gateway to multiple liquefied natural gas (LNG) terminals. Together, these infrastructure upgrades give Devon a direct highway to premium Gulf Coast pricing.
This pipeline capacity would not only accommodate all of Devon’s current gas production in the Delaware Basin, but also leave meaningful headroom for future growth.
Devon did not stumble into this moment. It spent years preparing for it. The 2021 acquisition of WPX Energy gave the company a massive 400,000-acre footprint in the Delaware Basin. From there, Devon methodically built midstream relationships, invested in gathering and processing, and co-funded critical pipeline capacity – all while retiring nearly $1 billion of debt and returning hundreds of millions of dollars to shareholders.
Its fundamentals are also solid. The company recently beat fourth-quarter earnings estimates. It performed exceptionally well on the production side, although it warned that several major winter storms would cut production for the first quarter.
Still, that hasn’t stopped the oil & gas company from outperforming the hyperscalers that require the kind of power that Devon provides.
The Roundhill Magnificent Seven ETF (MAGS), the ETF that tracks the Mag 7 companies, is barely up 2% this year. In contrast, Devon is up 14% year-to-date.
The bottom line is that AI-driven power demand is quietly rewriting the natural gas story, and Devon is positioned to thrive. In a market obsessed with the digital future, Devon represents something rarer: a hard-asset business, finally unshackled, standing at the center of a real-world demand shock.
Mag 7 Killer No. 2: A Turnaround Play in the Solar Sector
Our second Mag 7 killer makes the electrical balance-of-system (EBOS) products used in utility-scale solar and battery storage. That means the wiring, connectors, and power-delivery gear that carry electricity from the panels to the inverter.
Three positive developments at the company suggest its stock has become a timely buy.
First, Shoals recently announced a proposed merger with Array Technologies Inc. (ARRY), whose products complement Shoals’ own. Array, based in Albuquerque, New Mexico, makes solar trackers, the mounts that turn panels to follow the sun through the day. Over the past two years, Array has also bought APA Solar, which makes engineered foundations and fixed-tilt racking, and Affordable Wire Management, which makes cable-management gear. The industry calls these products the “structural balance of system,” meaning the steel that holds a solar farm up.
Shoals supplies the “electrical balance of system,” meaning the wiring that carries the power out. Together, the two companies can supply nearly every major structural and electrical part of a utility-scale solar project. The merger could also produce new products. For example, Array could design tracker frames with built-in mounting points for Shoals’ wiring, so field crews no longer have to drill holes on site.
Shoals and Array will combine in an all-stock deal that should close in the first quarter of 2027. Each Shoals share will convert into 1.6931 Array shares. Shoals holders will own about 65% of the combined company, and Shoals CEO Brandon Moss will run it.
Management targets $35 million of synergies in year one, rising to more than $100 million by year three, about two-thirds of it from extra sales. Moss points out that one extra gigawatt of tracker sales adds about $100 million of revenue and $30 million of EBITDA. That’s two or three projects. In his words, “We don’t have to make miracles.”
Second, the news from Washington D.C. Solar stocks rallied sharply after the midterms delivered a Democratic sweep of both chambers of Congress. With both chambers, Democrats will likely push to extend the business solar tax credits beyond 2027. Array also counts the Section 45X domestic-manufacturing credit in its margins, and a Democratic Congress is unlikely to try to repeal it. But President Trump still holds the White House, and neither chamber has anything close to a two-thirds majority. So the gain looks more like protection from new cuts than a fresh round of subsidies.
Third, the valuation gap. Nextpower is currently trading around $91.50. That price puts its enterprise value at about 14 times the midpoint of its EBITDA guidance for the fiscal year ending March 2027, nearly double the combined Array-Shoals multiple.
We do need to take a look at the risks. Array’s largest acquisition, its 2022 purchase of STI Norland, has gone poorly. STI’s revenue fell 83% in the first half of 2026. Array just cut its 2026 guidance, and its third-quarter gross margin fell to 22.8%, partly because of a one-time charge. The combined company will close with only about $30 million of cash. Guggenheim’s Joseph Osha called that “a little thin.” Both companies’ shareholders still have to approve the deal. And until it closes, Shoals stock will move in lockstep with Array’s.
But the reward outweighs those risks. Given the all-stock deal, the midterm results, and the valuation gap with Nextpower, I anticipate Shoals to be one of the bellwethers of the regime change.
Mag 7 Killer No. 3: The Largest American Aluminum Producer
If power is the blood circulating through data center infrastructure, metals are the bones.
Our third and final Mag 7 killer is a company digging up just that.
Every ton of metal pulled from the ground is a claim on the AI buildout. Unlike Software-as-a-Service (SaaS) vendors or chip designers, metals companies don’t need to guess which AI model wins or which agent framework dominates; they just need to deliver the raw materials that make the entire ecosystem possible.
Aluminum demand is also accelerating. Every high-voltage line that feeds an AI data hub consumes one to two tons of aluminum per megawatt delivered. Each new stretch of long-distance transmission deepens the world’s appetite for this versatile metal. From 104 million tons of demand in 2024 to an estimated 120 million by 2030, global aluminum consumption is set to grow almost as relentlessly as copper’s.
The aluminum market appears to have caught the scent of this rising demand. Prices have climbed roughly 10% year to date, reaching a new three-year high – welcome news for Alcoa Corp. (AA). After suffering a tariff-induced selloff earlier in the year, Alcoa’s shares have been trending higher, and I expect that uptrend to gain momentum – driven not only by firmer aluminum prices, but also by the company’s exceptional fundamentals.
Alcoa is not just the largest American aluminum producer; it is also among the world’s most environmentally progressive. Producing aluminum requires immense amounts of electricity, and that energy intensity is reshaping the industry. Increasingly, companies such as Tesla Inc. (TSLA) are seeking to source their aluminum from clean-energy smelters powered by hydro, nuclear, or renewables. That shift is elevating low-carbon producers like Alcoa and Norsk Hydro ASA (NHY) to the top tier of the aluminum world.
In 2025, renewable energy powered roughly 87% of Alcoa’s smelting operations and about 70% of Norsk Hydro’s. This alignment with the global push toward decarbonization gives both companies a durable strategic advantage, and positions them not merely as metal producers, but as critical enablers of the cleaner, more electrified world AI will depend on.
According to research from Metals Focus, a boom in demand for AI applications will create an echo-boom in demand for the high-specification semiconductors and sensors that enable AI technologies to operate optimally.
In the end, the market may reward not those who build the virtual world, but those who power it. The data revolution will always need its dreamers, but it will depend on the miners who turn electricity and metal into the blood and bones of artificial intelligence.
In the race for AI supremacy, the hyperscalers may scorch their balance sheets, but the miners will still be cashing the checks. While hyperscaler shareholders wrestle with wafer-thin cash cushions and swelling debt, metals firms operate with clearer economics. Their business models are not theoretical; they are measurable and proven.
They don’t need to care whether GPT-7 outperforms Gemini Ultra or whether OpenAI’s next model hallucinates less than Anthropic’s. They get paid every time a new data hall lights up, every time another transformer hums, every time a ton of aluminum vanishes into a conduit.
Therefore, for investors seeking exposure to the AI revolution without betting on which version of intelligence wins, miners offer a compelling opportunity.
renzy.
A small portion of the coming regime change has already taken place. Devon’s, Shoals’ and Alcoa’s killer gains over the Mag 7 prove just that.
Regards,
Eric Fry