The AI Opportunity Is Bigger Than a Ticker

At Future Proof Citywide, the most useful conversations were not about whether AI is “real” or “hyped.” They were about where the durable opportunities are hiding, how advisors can explain them, and why a narrow view of technology investing is probably missing the point. My conversation with Evan Feagans, Managing Director at TCW, made that especially clear. His perspective is rooted in a broad mandate: co-managing seven tech-focused thematic equity strategies with more than $2 billion in assets, including the TCW Artificial Intelligence ETF, AIFD, while also serving as TCW’s senior equity analyst covering internet and media across all equity strategies.

What stood out most was how disciplined his framing was. Feagans is not approaching AI as a single-stock story, and he is not treating it like a magic label that can be slapped onto any company with a press release. His view is that technology leadership still matters most, but that leadership will not be evenly spread across the market. As he put it, “the value only accrues to a couple companies.” That matters for advisors because it pushes the conversation away from generic tech exposure and toward a more selective, research-driven approach.

The other important point is that he is clearly thinking about AI as an ecosystem, not a theme in isolation. In his view, the investable universe is widening fast, and that widening does not just create more names to own; it creates a new way to think about where value is captured. He described the landscape in three layers: “the AI enablers, the semiconductors, things that go into an AI data center,” then “the AI systems,” and finally “AI adopters.” That framework is useful for advisors because it helps explain why the story is larger than the Magnificent Seven and broader than a simple “chips and cloud” narrative.

Why selectivity matters

Feagans’ comments on prior technology cycles should resonate with anyone who has lived through more than one market mania. He pointed to the dot-com era as a reminder that even when a category is transformative, not every participant survives the cycle. His example was blunt: if you owned Amazon through that period, you did very well; if you owned a basket of speculative IPOs, “you lost your shirt.” AI does not have to repeat the dot-com bubble in a neat, one-to-one way for the comparison to be useful. Innovation can be massive while the investable outcomes remain highly concentrated.

That is where his focus on business quality comes in. Feagans said TCW looks for “good business models, scalable business models, good unit economics,” and companies that “ultimately do things better and cheaper than their competitors.” That shifts the AI discussion away from who has the loudest story and toward which companies can compound value through product strength, efficiency, and economics that hold up over time.

It also gives the conversation something more concrete than a theme. A technology strategy is easier to understand when it is tied to business fundamentals rather than excitement alone. The question is not simply whether AI is important. It is which companies can monetize it sustainably.

Internet and media overlap

One of the more revealing parts of the conversation was how naturally Feagans connected his coverage of internet and media companies with the AI strategies he helps manage. The overlap is not accidental. “Most of the companies, at least on the internet side, are well within the AI universe and are actually driving it,” he said.

That is how thematic investing often works in practice: not as a clean sector silo, but as a web of related businesses. Meta, Amazon, and Google are not just beneficiaries of AI capital spending. “If you think about Meta, Amazon, Google, those are some of the largest spenders, the biggest drivers, of the AI technology super cycle,” Feagans said. That helps explain why some of the most familiar names in portfolios may still be central to the next phase of innovation.

His media examples also pulled the conversation out of abstraction. Feagans talked about AI being used for special effects, early script work, game asset generation, and other forms of content creation that can improve workflow without replacing the entire creative process. He was careful not to overstate the point. “I ultimately don't necessarily agree that we'll be watching 100% AI-generated content,” he said. “I think there's a limit to what consumers will tolerate.”

That distinction is important. AI may not eliminate existing media models so much as reshape how they operate, how much they can produce, and how efficiently they can do it.

The opportunity under the surface

Feagans’ practical message was that the AI opportunity is bigger and more diversified than many investors realize. He noted that some people assume they already have enough exposure through the Magnificent Seven, but he argued that “the opportunity is so much larger and so much more diversified than that.”

He backed that up with a striking view of the infrastructure buildout. A couple of years ago, Feagans said, expectations were for hyperscalers like Google, Meta, Amazon, and Microsoft to spend roughly $200 billion on AI infrastructure this year. “It’s going to look more like $500, $600 billion now,” he said. “Every quarter these things get revised higher.” He described the market as moving toward “a trillion dollars of investment” and suggested that even that may not be the ceiling.

The point is not to anchor on one number. It is to understand the scale of the industrial buildout happening underneath the AI story. That gives the theme staying power beyond short-term market enthusiasm and helps explain why a focused AI strategy is not simply a repeat of owning the largest platform stocks.

Feagans also highlighted the importance of AI enablers such as networking, memory, custom chips, and interconnects. Those are not always the most visible names, but they are essential to how the infrastructure scales. As he put it, “there’s just a lot more under the surface too, that people don’t necessarily own at this point.”

Media gets reworked

The media side of his coverage added another layer to the AI story. He pointed to social platforms as an early example of how machine learning can change the economics of a business, not just make it more efficient. TikTok was the clearest example. “You didn’t have to follow any accounts. You didn’t have to do anything,” he said. “You just had to scroll,” and the algorithm would eventually find the specific interests that kept people engaged.

Meta’s response showed how quickly those dynamics could reshape a platform business. Feagans said the company moved away from relying only on the “pure knowledge graph,” or what users followed on Instagram and Facebook, and embraced the kind of machine-learning algorithm TikTok had pioneered. “That’s really what reversed that trend on the revenue side,” he said.

That shift is now showing up in advertising. Feagans described Google’s Performance Max and Meta’s Advantage+ as AI targeting products that made sophisticated advertising tools available to a much wider range of businesses. “It let a restaurant down the street get similar results in targeting to what P&G would get,” he said. In his view, that increased return on ad spend and helped accelerate revenue for those platforms.

The same logic applies to content. AI may lower production costs and give individual creators access to tools that once required larger teams, from effects and editing to music and asset generation. That does not mean media companies become less important. It may mean their role changes. As Feagans put it, “that makes content curation all the more important.”

If AI leads to more content, more personalization, and more niche creation, then discovery becomes even more valuable. The winners may not simply be the companies that generate the most content. They may be the ones best able to surface what people actually want to watch, hear, read, or share.

Beyond the AI trade

AI is not one trade, one ticker, or one group of mega-cap names. Feagans’ larger point was that the opportunity is broad, but not evenly distributed. Some companies will benefit directly from infrastructure spending. Some will benefit from software adoption. Some will use AI to defend or expand their existing businesses. Others will be left behind.

That is why his emphasis on selectivity matters. Across our conversation, Feagans kept returning to business quality, scalable models, unit economics, and the ability to do something better or cheaper than competitors. His role across seven thematic strategies gives him a broad view of the technology landscape, but his answers were notably specific. He was not selling a slogan. He was describing a process.

That process is especially important in a market where almost every company now has an AI story to tell. The harder work is separating exposure from value creation. AI is already reshaping how companies build products, spend capital, target customers, and create content. The opportunity is not just in owning the headline names. It is in understanding the full stack, the business models that can scale, and the companies that can turn technological change into durable economics.

To learn more about TCW’s technology and thematic equity capabilities, visit the TCW website.