Listen to the audio version of this article (generated by AI).
Hello, Reader.
In machine learning – the type of AI that teaches computers to learn from examples – not all data is created equal.
“Hurtful data,” be it mislabeled, misleading, or biased, can degrade AI models. That’s like a photo of a cat being labeled “dog.” Other data is “useless” because it is repetitive, low quality, or adds no meaningful new information.
If the examples are wrong, the model learns the wrong patterns.
That means AI progress is becoming less about who has the biggest model and more about who has the best data.
Companies that apply AI are not created equal, either.
I’ve been talking about these “AI Appliers” here at Smart Money. These companies are adopting AI technologies to boost efficiency, productivity, and profitability.
Many could profit enormously as they deploy AI technologies throughout their operations and potentially see gains not unlike the companies that used internet infrastructure in the dot-com era and went on to dominate the global economy.
But there’s a catch…
Not every company using AI has an advantage. We’ve reached the stage in the technological cycle where execution matters more than adoption.
That is especially true as “A-AI” – AI models that can act all on their own – becomes more prevalent
In today’s Smart Money, I’ll look at a company that highlights the limits of legacy AI Appliers struggling to adapt as A-AI accelerates. Then, I’ll break down an AI Applier that can successfully leverage A-AI to strengthen its core business and gain a competitive advantage.
Let’s dive in…
A Legacy AI Applier Struggling to Keep Up
Xerox Holdings Corp. (XRX) is a well-recognized company built around printers, copiers, and traditional office services
In the early 2010s, management wisely anticipated a decline in its traditional printing business. And so, they acquired a company called Affiliated Computer Services to offer “business process outsourcing,” also known as “BPO.” This new business offered IT outsourcing, administration, call-center, HR and other back-office services to complement its legacy business.
However, the strategy did not solve Xerox’s long-term problem: print volumes were still going down, and BPO was a low-margin, complex business that faced intense competition from foreign firms. And so, in 2016, Xerox sold the BPO division, only to jump in again in 2023 after its printing business saw further declines.
This time, they tried a narrower “workplace technology” strategy.
Things looked promising at first. Revenues stabilized… and even grew slightly in 2025 thanks to acquisitions and the company’s entry into AI-enabled automation. Insiders at Xerox were aggressively buying stock in their own company as late as April 2025.
But things began to fall apart again this year as A-AI started to take hold.
You see, Xerox’s latest foray into outsourcing involves businesses that A-AI can increasingly automate. Areas like:
- Document workflow automation.
- Mid-market IT services
- Knowledge-worker productivity services
And Xerox’s inconsistent BPO history means it doesn’t have the quality data needed to fight back.
In addition, Xerox’s acquisitions mean the firm holds over $4 billion in debt (more than 10 times its market capitalization). A-AI is shrinking profits in Xerox’s “growth” business right as the company needs it most.
Of course, management is continuing to fight back. Last week, Xerox released Xerox IT as a Service (ITaaS), an AI-powered operations platform designed to simplify IT for companies. Basically, businesses will pay Xerox to manage and support their technology systems instead of handling IT themselves.
But insiders are no longer buying shares like they did in 2025. A-AI competition is becoming clearer by the day, and Xerox might not have the data or financial strength left to beat back the competition.
Xerox may apply AI, but that doesn’t mean it’s guaranteed a long, or successful, life as an AI Applier.
Here’s a better “match” instead…
A Match Made for the AI Applier Era
Match Group Inc. (MTCH) is the undisputed titan of online dating websites and apps, with roughly 82 million monthly active users – about 15 million of whom pay for subscriptions – and an estimated 30–40% global market share.
Its portfolio includes Tinder, the world’s No. 1 dating app, along with more than a dozen other online dating brands like Hinge, OkCupid, Plenty of Fish, and Match.com.
While these consumer-facing initiatives are critical to user growth and retention, Match is also using AI to transform its internal operations, making the company more efficient and more collaborative.
For example, Match has deployed AI coding assistants like “Cursor” globally to speed up development cycles and reduce engineering workloads.
Additionally, nearly 1,000 engineers now work through a shared GitHub system with AI tools, which lets teams see each other’s code, reuse good features, and build products faster and more consistently – similar to larger tech firms like Meta Platforms Inc. (META) and Microsoft Corp. (MSFT).
Match has also created a centralized AI tooling group that builds and maintains shared AI infrastructure. This approach gives smaller brands in the company’s portfolio, like HER or The League, access to the same advanced capabilities as the flagship apps without having to duplicate engineering efforts.
By embedding AI at multiple touchpoints – from onboarding and match recommendations to internal product development – Match is building a unified ecosystem where data, technology, and human creativity reinforce each other.
This dual focus on consumer experience and operational efficiency is not simply an experiment; it is the strategic foundation for the company’s revitalization plan.
Match Group could actually benefit from the rise of A-AI if it becomes the company that owns the “AI relationship assistant” layer for dating.
Unlike Xerox, Match’s core business is already built around recommendations, personalization, and matching – exactly the kinds of problems A-AI is good at solving.
Instead of replacing Match’s business, A-AI could make its products more useful. Applying the technology would be a boon, rather than a bust.
And in a world of A-AI, execution beats adoption. The right AI Applier isn’t just using the technology; it’s being made stronger by it.
That’s why the gap between AI Appliers is widening.
And for us investors, that difference is becoming the entire story. It’s likely to separate the long-term winners from the rest.
I share more about the A-AI shift – and the Applier companies set to profit from it – in my newest, special broadcast. I also give away my number-one AI Applier stock pick – absolutely free.
Regards,
Eric Fry
