The venture capital landscape is currently undergoing its most significant transformation since the dawn of the internet. As artificial intelligence moves from theoretical research to the bedrock of global enterprise, the rules of investment, valuation, and market defensibility are being rewritten in real-time.
Last week, in the sun-drenched innovation hub of El Segundo, California, TechCrunch’s StrictlyVC event convened two of the industry’s most perceptive voices to dissect this shift. Carter Reum, co-founder of the $2.5 billion firm M13, and Chang Xu, a partner at Basis Set Ventures—a firm that staked its reputation on AI long before it became a household acronym—offered a sobering yet optimistic look at a market moving at breakneck speed.
The State of the AI Infrastructure Bubble: Paradoxical Growth
The central question looming over every board meeting in Silicon Valley and Los Angeles is simple: Are we in an AI bubble?
According to Chang Xu, the answer is a complex "yes and no." She argues that while valuations are undeniably high, they are often tethered to growth metrics that defy historical precedent. "ChatGPT went from zero to $40 billion in revenue in six months—that is unprecedented," Xu noted. She highlighted a portfolio company, OpenArt, which scaled from $1 million to $70 million in Annual Recurring Revenue (ARR) in just two years while remaining largely cash-flow positive.
For investors like Xu, the "bubble" narrative ignores the reality of compounding, hyper-accelerated growth. If a startup can reach such scale with a lean team of 20, traditional valuation multiples become obsolete. However, she warned that applying these aggressive metrics across a broad portfolio is a recipe for disaster.
Carter Reum offers a more historical perspective. He compares the current AI frenzy to the advent of the cloud, the iPhone, and even the automotive revolution of the 1920s. While he acknowledges the cycle is steeper and faster, he identifies a critical difference: the competitive landscape. "In past cycles, innovators competed with other innovators," Reum explained. "Today, startups are competing with the largest, most well-funded tech companies on the planet. For the first time in history, the incumbents—Google, Microsoft, Meta—have the distinct advantage of data, capital, and talent."
Strategic Frameworks: Investing Above and Below the AI
How does an investor price a deal when revenue sustainability is obscured by the volatility of the AI frontier?
Reum advocates for "cocktail napkin math." If a startup is selling AI software to brands, the firm must calculate whether the market size truly justifies the cost. If the math doesn’t add up—if the value proposition doesn’t promise a 3x or 5x improvement in efficiency or revenue—M13 walks away.
Xu’s firm, Basis Set Ventures, utilizes a more granular framework, focusing on investing "below" and "above" the AI layer:
- Below the AI: This involves infrastructure—databases, version control, and deployment tools—that were originally built for human developers. As AI agents begin to utilize this infrastructure, it requires a total re-architecture. Xu noted that demand for "GitHub for agents" has surged, presenting a massive opportunity for foundational tools.
- Above the AI: This is where defensibility is paramount. In crowded markets, Xu looks for long-term differentiation that remains intact even as the underlying models evolve.
Navigating the "Hyperscaler" Threat
One of the most pressing concerns for founders today is being "steamrolled" by incumbents like OpenAI or Google. Reum suggests that the best defense is "friction as a moat."
"We love regulated industries," Reum stated. By targeting sectors like 911 call center technology or complex healthcare systems, startups can build defensible niches. While hyperscalers might eventually enter these spaces, the regulatory hurdles and the sheer complexity of integration provide smaller firms the time they need to reach scale.
Depth Markets vs. Velocity Markets
The panel introduced a critical distinction in investment strategy:
- Velocity Markets: Where fast followers can overtake leaders through pure speed. Here, execution is the only metric that matters.
- Depth Markets: Where the "hard things remain hard." Xu cited a portfolio company utilizing transgenic chickens to manufacture complex proteins. No matter how fast an AI model is, the biological reality of the process remains a fixed, long-term barrier to entry. This is the hallmark of a depth market.
The L.A. Ecosystem and the SpaceX Effect
The conversation inevitably turned to the geographical shifts in tech. The impending SpaceX IPO is expected to inject unprecedented liquidity into the Los Angeles economy. Unlike previous liquidity events, which were often concentrated among a small group of institutional investors, the SpaceX windfall will be distributed across a vast network of employees and local stakeholders.
Reum believes this will trigger a "second wave" of innovation in Southern California. "The first wave of AI is technical," he explained. "Technical talent is currently concentrated in places like San Francisco. But what comes after the technical wave? New business models, creative thinking, and cultural integration."
This is where Los Angeles holds a distinct advantage. While San Francisco excels at the "how" of AI—the compute, the engineering, the optimization—L.A. excels at the "why." As Xu succinctly put it: "The next frontier in AI isn’t more compute; it’s taste."
Implications: The Second and Third Ripples
Looking toward the future, Reum encouraged investors and founders to adopt a dual-vision approach: "You need a microscope in one eye and a telescope in the other." The microscope focuses on the brutal, day-to-day execution required to survive, while the telescope identifies the shifts that could render a business model obsolete overnight.
We are currently in the "first wave," which is often the most crowded and obvious. The true returns, according to the panel, will lie in the "second and third ripples"—the business models we cannot yet imagine.
The story of venture capital is, and has always been, a story of bad ideas becoming good again. Whether it is the rise of creative AI tools for Hollywood or the emergence of new coding paradigms like Cursor, the market is constantly proving that the most successful ventures are those that iterate at the edge of the possible.
As the industry matures, the focus will likely shift from the raw power of Large Language Models to the nuanced application of those models within specific cultural and creative domains. For Los Angeles, a city defined by its ability to blend narrative, brand, and influence, the AI era may be the catalyst that cements its position as the world’s leading hub for the "creative-technical" synthesis.
The message to the next generation of founders is clear: Stop looking for the most obvious path. Find the depth market, embrace the friction, and start building for the world that will exist three years from now, not the one that exists today.
Suro Senen
Iffa Jayyana
