Is OpenAI devouring the application layer? a16z: Opportunities lie beyond the "Yellow Brick Road," and the best is yet to come for founders

a16z partners point out that the AI application layer is not a single battlefield; startups should avoid horizontal tools directly targeted by large model companies and instead pivot to deep cultivation in vertical industries. This article is sourced from a Twitter post. (Context: Google leads investment in AI routing platform OpenRouter, valuation reaches $1.3 billion, growing 240% in one year) (Background: Sam Altman discusses with a16z founder: OpenAI will aggressively bet on infrastructure, Sora is a key strategic tool) This is precisely the question a16z partner Joe Schmidt attempts to answer in this article. Using the "Yellow Brick Road" from *The Wizard of Oz* as a metaphor, he divides AI application opportunities into two categories: one is the main road that large model companies are entering themselves, such as code generation, writing, image generation, general-purpose agents, and horizontal office assistants; the other is "the rest of Oz," which refers to vertical scenarios that delve deep into industrial processes, rely on complex workflows, data accumulation, compliance governance, and system integration capabilities. In his view, the real opportunity for startups lies in the latter. From sales to insurance, Joe Schmidt repeatedly emphasizes the same logic: what enterprises are truly willing to pay for is not a smarter chat window, but a system that can take responsibility for business outcomes. It needs to understand the messy state of customer data, handle multi-person approvals and edge cases, assume compliance and audit responsibilities, and also handle migration, routing, and cost optimization for customers as models continuously upgrade. This is also the core judgment of this article regarding the next generation of enterprise software: underlying models will become increasingly powerful and interchangeable; but what is truly irreplaceable is the data, processes, governance capabilities, and operational memory accumulated around specific industries and workflows. The opportunity for AI application companies does not lie in competing with model companies for the "Yellow Brick Road," but in walking into those places that are more complex, dirtier, slower, but also closer to real business value. Recently, I have been hearing the same question from founders and potential employees: Is there anything left to do in the AI application layer? Or will OpenAI and Anthropic eventually kill everything? Behind this question lies a typical AI-style anxiety. Some have already concluded: if you don't want to be relegated to the permanent bottom layer, the only position with long-term value is either inside a large model lab or starting a business in robotics, hard tech, or similar frontier fields—theoretically, doing things that "labs can't touch." Because if every category of software is going to be swallowed, either having its corresponding work absorbed by Codex or Claude, or becoming unnecessary due to some future model, then the best choice seems to be: run! I admit that I am almost an AI maximalist myself, and I think they are half right. Large model labs are indeed entering large swaths of the application layer. But the "application layer" is not a homogeneous collection of opportunities. The truly critical criterion for judgment is: are you walking the "Yellow Brick Road," or are you in the rest of Oz? Note: The "Yellow Brick Road" is the main path in *The Wizard of Oz* that leads to the heart of the Emerald City to meet the "Wizard." The so-called "Yellow Brick Road" is the path we use to describe what large model labs are walking and investing huge resources into. Problems like code generation, writing, and image creation are naturally suited for labs because they get better as the model's raw capabilities improve: every dollar invested in pre-training and post-training directly improves product quality. But the rest of Oz contains more complex and usually more vertical problems. They cannot be solved simply by providing an enterprise user with a horizontal tool that connects to standard tools and computer operation capabilities. The value here comes more from the scaffolding around the model: this scaffolding makes the output credible and compliant in specific industries and allows it to truly enter business processes. The raw capability of the underlying model is of course still critical, but it is no longer everything. We are seeing this in real-time. OpenAI and Anthropic are effectively admitting to the market: they cannot solve all problems with a general-purpose AI colleague. They have already announced investments in large-scale front-line deployment joint projects, building complete companies around configuring and customizing models for enterprises. If they truly believed that the next model release would solve these problems, they wouldn't be pouring billions of dollars into these types of projects. So, if you want to make money by building AI applications, don't walk the Yellow Brick Road; go build in the rest of Oz. Below are some lessons learned in practice by us and some founders in our portfolio. If you are going to start a company, the Yellow Brick Road is the most visible path, but it is also the most dangerous. Take a high-performance model