AI Boom: Market-First Investing in Infrastructure and Content Creation

The AI Boom: Navigating the Early Days of a Technological Supercycle

This conversation with Martin Casado, a General Partner at a16z, offers a compelling perspective on the current AI boom, framing it not as a fleeting trend but as an early-stage phenomenon akin to the mid-1990s internet era. The non-obvious implication is that the immense value being created is still largely unrecognized, with significant opportunities for those who understand the market dynamics rather than just chasing company-specific narratives. Investors, founders, and technologists seeking to capitalize on this wave will benefit from adopting a market-first investment lens and recognizing that true competitive advantage often lies in patiently identifying and backing emergent leaders in rapidly expanding, yet still nascent, markets. Those who dismiss the current energy as mere speculation risk missing what Casado believes could be the greatest supercycle in two decades.

The 1996 Moment: Why the AI Boom Has Room to Run

The current fervor around Artificial Intelligence echoes the nascent days of the internet in the mid-1990s, according to Martin Casado. This isn't just a rehashing of past tech cycles; it's a strategic observation that the AI revolution is still in its early innings, offering a unique window for value creation. Casado’s investment philosophy, honed over decades as a founder and investor, emphasizes a market-first approach. Instead of evaluating companies in isolation, he focuses on identifying burgeoning markets and then finding the best founder to lead within that space. This perspective is crucial because, as he notes, "the market creates the company in most cases, not the other way around."

The energy surrounding AI today mirrors the optimism and chaos of the late 90s. Casado recalls the palpable excitement of that era, where computer science was a niche discipline that exploded with the advent of the web. Now, AI is experiencing a similar "slow boil that erupted." However, he distinguishes the current AI boom from the dot-com bubble of the early 2000s. While valuations are "spicy," they are underpinned by real business usage, revenue, and sustainable growth metrics, unlike the speculative ventures of the past that often lacked revenue. The capital required for foundational AI models, for instance, is immense, signaling a shift towards capital-intensive, revenue-generating businesses. This is not a speculative frenzy; it's the market normalizing around genuine value creation.

"The current technology wave is you can actually deploy capital and you can get revenue on the other side of it and i think that is what the market is trying to normalize but there's a true value being created in this ai and i think that if money's not following it it's going to miss the greatest super cycle in the last 20 years."

-- Martin Casado

The distinction between the current AI wave and the 2021 exuberance is critical. Casado argues that 2021’s excitement was driven by a COVID-induced influx of capital and macro trends, not sustainable business reasons. AI, on the other hand, has a foundation built on years of development, with demonstrable retention, growth, and margins. This makes the current phase feel more akin to 2019, with a significant runway ahead. The challenge for investors, and indeed for the industry, is to move beyond the hype and understand where real, durable value is being created. This requires patience and a willingness to wait for market leaders to emerge, a strategy that, while not perfect, yields results superior to market norms.

The Hidden Costs of "Solving" AI and the Power of Pro-Social Behavior

While the AI boom presents immense opportunities, Casado highlights a critical gap between enterprise aspirations and realized value. Many companies are scrambling to implement AI projects, often driven by board mandates rather than clear strategic vision, leading to high failure rates. He posits that AI, in its current form, is primarily an individual "prosumer" technology--tools like ChatGPT or Cursor used by individuals. The value for organizations lies in their employees leveraging these tools. Direct enterprise sales for AI are still nascent. Casado’s advice for enterprises? Partner with vendors and product companies that are successfully implementing AI, rather than attempting internal projects prematurely. This mirrors the early internet days, where direct enterprise adoption lagged behind individual use.

"The problem with that is like composition is just real. We've got an existing, you know, set of tool, we've got existing tool chains that use like components of outputs that you're going to want to use. And so I think that's a mistake."

-- Martin Casado

Casado also touches upon the debate between monolithic "god models" and the composition of smaller, specialized models. He believes both paths are viable and that assuming one will dominate is a mistake. The "bitter lesson" argument suggests that larger, more capable models will continue to advance, but the reality of existing toolchains and the desire for fine-grained control will also drive the use of composed models. This highlights a systems-level understanding: the AI ecosystem is not a single entity but a complex interplay of different approaches.

Furthermore, the conversation delves into the critical role of pro-social behavior in driving innovation. Casado recommends "The Weirdest People in the World" by Joe Henrich, which explores how historical shifts, like the Protestant Revolution and the rise of free markets, fostered pro-social behavior necessary for cooperation and planetary-level innovation. This is crucial because, as he notes, the ultimate enemy is entropy, and overcoming it requires collective action and trust. The idea that humanity is on a continuum of improvement, as explored in Fukuyama's "The End of History and the Last Man," suggests that while we may not have "arrived," the trajectory is one of increasing sophistication in how we interact and solve problems. This long-term perspective is vital for understanding the enduring potential of AI and other technological advancements.

The AI Coding Revolution and the Quest for 3D Representation

The impact of AI on coding has been particularly surprising for Casado, a lifelong developer. He notes that AI has made programming accessible again, removing the burden of learning constantly evolving frameworks. Tools like Cursor, which he is involved with, are transforming the developer experience by integrating AI directly into the IDE. The market for developer tools, he estimates, is in the trillions of dollars, making it an infinitely sized market where AI is proving exceptionally effective.

"If you ask me what is the one area that ai has surprised you it's in coding. i have listen i've been developing my whole life and i would never have guessed it'd be this good."

-- Martin Casado

Defensibility in this rapidly fragmenting market, Casado argues, doesn't come from AI itself, which he believes is largely commoditized. Instead, it lies in traditional modes: building two-sided marketplaces, integration workflows, or creating new behaviors that incumbents struggle to adopt. The immediate advantage for AI companies lies in overcoming customer acquisition challenges due to the technology's "magic."

Another area of profound impact is 3D content creation, exemplified by World Labs, a company Casado is closely involved with. The ability to generate realistic 3D scenes from 2D images has far-reaching implications for virtual environments, design, AR/VR, and critically, robotics. Embodied AI, which needs to understand depth and spatial relationships, relies heavily on 3D representation. By bringing the marginal cost of 3D content creation to zero, markets that were previously non-existent, like VR content, can explode. This is a foundational problem that, when solved, unlocks immense value, much like the marginal cost of image creation did for companies like Midjourney.

Key Action Items

• Adopt a Market-First Investment Lens: Prioritize identifying rapidly growing markets before selecting individual companies. This approach, favored by Casado, leads to more robust long-term investment decisions. (Immediate)
• Focus on Pro-Social Behavior: Recognize that overcoming complex challenges, whether in AI development or broader societal issues, requires trust, cooperation, and pro-social dynamics. Seek to understand and foster these elements in your work. (Ongoing)
• Partner for Enterprise AI Adoption: For organizations struggling with internal AI projects, collaborate with specialized AI vendors and product companies that are demonstrating success. (Immediate)
• Embrace AI for Coding Productivity: Leverage AI-powered coding tools to enhance efficiency and focus on fundamental problem-solving rather than framework-specific minutiae. This offers immediate personal and team-level benefits. (Immediate)
• Invest in 3D Content Creation Capabilities: As AI democratizes 3D content, explore opportunities in VR, AR, robotics, and embodied AI where realistic 3D representation is a key enabler. This is a longer-term investment thesis with significant upside. (12-18 months)
• Distinguish AI Hype from Sustainable Value: Critically assess AI initiatives, differentiating between speculative trends and those with clear revenue generation, user adoption, and sustainable growth metrics. (Immediate)
• Advocate for Innovation-Friendly Policy: Support policies that foster technological advancement and open-source development, recognizing the long-term national and global implications of innovation leadership. (Long-term investment)