Raphael Schaad, a notable figure in the tech industry, recently highlighted a significant opportunity for AI startups, emphasizing that specialized "last mile" applications can flourish despite the rapid advancements of large frontier AI labs such as OpenAI, Anthropic, and Google. Schaad's perspective, shared in a widely circulated tweet, directly addresses a common concern among emerging founders regarding the potential market dominance of these tech giants. He posits that the overall market for AI solutions is expanding, not contracting, creating ample room for innovative ventures.
"Won’t the frontier labs just do everything?" Schaad quoted an MIT student's question, acknowledging the prevalent apprehension within the startup community. He asserted that the true value for new companies lies in deep domain expertise and direct customer engagement, enabling them to bypass direct competition with general-purpose AI offerings from larger entities. This strategy focuses on delivering specific outcomes rather than generic tools.
Schaad provided compelling examples of companies successfully employing this niche strategy. Graze Mate, a Y Combinator W26 startup, is reportedly developing cattle-herding drones, showcasing a highly specialized application distinct from the broad capabilities of major AI labs. While specific details on Graze Mate are not widely available, the concept exemplifies a focused, industry-specific AI solution. Similarly, the tweet referenced "Lance" for integrating AI with dental insurance verification systems, a sector where AI is indeed making significant inroads by automating complex administrative tasks, as evidenced by numerous companies in the field.
Another example cited was Milliray, which is said to navigate NATO procurement processes using AI. While direct information about a company named "Milliray" specifically for NATO procurement is scarce in public records, the concept underscores the potential for AI to streamline intricate, high-stakes governmental and defense logistics. These examples collectively illustrate the power of applying AI to highly specific, often overlooked, industry challenges.
According to Schaad, the AI market is undergoing substantial expansion, not merely cannibalizing existing software spending. He suggested that AI is unlocking an additional 5-6% of the market that was previously handled by human labor, representing a much larger opportunity for new startups focused on delivering complete outcomes rather than just tools. This shift allows for the creation of "full-stack" AI companies that function as specialized services, such as AI law firms or accounting firms.
Conversely, Schaad warned against ventures that offer only a "thin UI layer on top of ChatGPT with no domain expertise" or general-purpose chatbots, deeming them high-risk. Such products, he noted, are vulnerable to obsolescence as core model capabilities improve. Instead, defensible strategies involve building tools for specific industries or creating comprehensive AI-driven services that provide a complete solution, where the customer seeks an outcome rather than just an interface.
