Google VP Warns of AI Startup Shakeout

Loading the player…

The generative AI boom fueled a rapid pace of startup creation. However, as the initial excitement subsides, two previously popular business models—LLM wrappers and AI aggregators—are increasingly viewed as warning signs.

Darren Mowry, who leads Google's global startup organization across Cloud, DeepMind, and Alphabet, suggests that startups relying heavily on these models have their "check engine light" illuminated.

LLM wrappers are essentially startups that build a product or user experience layer on top of existing large language models like Claude, GPT, or Gemini to address a specific need. An example would be a startup using AI to assist students with their studies.

"If your strategy primarily depends on the backend model doing all the heavy lifting, and you're essentially white-labeling that model, the industry is losing patience with that approach," Mowry stated on this week's episode of Equity.

Wrapping "very thin intellectual property around Gemini or GPT-5" indicates a lack of differentiation, according to Mowry.

He emphasized that for a startup to "progress and grow," it needs "deep, wide moats that are either horizontally differentiated or deeply specific to a vertical market." Examples of LLM wrappers with substantial moats include Cursor, a GPT-powered coding assistant, and Harvey AI, a legal AI assistant.

Save up to $300 or 30% on TechCrunch Founder Summit

Join 1,000+ founders and investors at TechCrunch Founder Summit 2026 for a full day dedicated to growth, execution, and practical scaling. Learn from founders and investors who have shaped the industry. Connect with peers at similar growth stages. Leave with actionable tactics you can implement right away.

Offer ends March 13.

Save up to $300 or 30% on TechCrunch Founder Summit

Join 1,000+ founders and investors at TechCrunch Founder Summit 2026 for a full day dedicated to growth, execution, and practical scaling. Learn from founders and investors who have shaped the industry. Connect with peers at similar growth stages. Leave with actionable tactics you can implement right away.

Offer ends March 13.

In short, startups can no longer expect to simply overlay a user interface on top of a GPT model and gain significant product traction, a strategy that might have worked in mid-2024 with the launch of OpenAI's ChatGPT store. The current imperative is to build sustainable, intrinsic product value.

AI aggregators are a specific type of wrapper—startups that combine multiple LLMs into a single interface or API layer to route queries across different models. These platforms typically offer an orchestration layer with monitoring, governance, or evaluation tools. Examples include AI search startup Perplexity or developer platform OpenRouter, which provides access to multiple AI models through a unified API.

Although several of these platforms have gained users, Mowry's advice for new startups is unequivocal: "Stay out of the aggregator business."

He explains that aggregators are generally struggling to grow or advance because users desire "some built-in intellectual property" to ensure they are directed to the most suitable model based on their specific needs—not due to underlying computational or access limitations.

With decades of experience in the cloud industry, including roles at AWS and Microsoft before joining Google Cloud, Mowry has witnessed this cycle before. He draws a parallel to the early days of cloud computing in the late 2000s and early 2010s as Amazon's cloud business surged.

During that period, numerous startups emerged to resell AWS infrastructure, positioning themselves as user-friendly gateways offering tooling, consolidated billing, and support. However, as Amazon developed its own enterprise tools and customers became proficient in managing cloud services directly, most of these intermediary startups were marginalized. Only those that added genuine value through services like security, migration, or DevOps consulting survived.

Today's AI aggregators face similar margin pressures as core model providers expand their own enterprise feature sets, potentially making middlemen redundant.

Conversely, Mowry is optimistic about developer platforms and "vibe coding," which had a landmark year in 2025. Startups like Replit, Lovable, and Cursor (all Google Cloud customers, according to Mowry) attracted significant investment and user adoption.

Mowry also anticipates robust growth in direct-to-consumer technology, where companies empower end-users with powerful AI tools. He highlighted the potential for film and TV students to utilize Google's AI video generator, Veo, to visualize their stories.

Looking beyond AI, Mowry identifies biotech and climate tech as significant growth areas, driven by both venture capital investment and the "incredible amounts of data" now available to startups, enabling them to create substantial value "in ways previously unimaginable."