Are small AI-based businesses sustainable, or will big tech make them obsolete?

Small AI-based businesses are considered sustainable by experienced founders, provided they move beyond being simple "wrappers" and focus on deep verticalization, proprietary distribution, or "AI-native" services. While big tech controls the underlying models, creators argue that small teams now have a historic advantage in speed and cost that allows them to challenge even the largest incumbents.

The Collapse of the Incumbent Moat

In the previous era of SaaS, massive incumbents were protected by "millions of lines of code built over decades." According to Y Combinator, AI has "collapsed the cost of producing software by 100x or more," effectively destroying the moats of legacy software giants Y Combinator — SaaS Challengers @ 00:00. This allows small teams of 2–3 people to build enterprise-grade products in months rather than years, enabling them to close multi-million dollar deals with Fortune 10 companies that would have previously ignored them Y Combinator — Startups That Sell to the Biggest Companies in the World @ 00:00.

Building Durable Moats in the AI Era

To avoid being made obsolete by big tech updates, founders must build "structural moats" that go beyond the AI model itself. Greg Isenberg notes that while it has "never been easier to build a million-dollar business," it has also "never been easier to lose all your equity value" if you don't have a sustainable advantage like a data or network effect Greg Isenberg — If I wanted to build $1M+ AI startup in 2025, I'd do this @ 00:00.

Key strategies for sustainability include:
* Verticalization: Instead of selling software to an industry, become the service provider. Y Combinator suggests that "regulated industries have higher expectations... that raises the bar and the moat for founders" Y Combinator — How to Build an AI-Native Services Company @ 00:09. For example, rather than building software for insurance companies, founders should aim to be the full-stack insurance carrier Y Combinator — How To Pick A Startup Idea @ 06:06.
* Proprietary Distribution: Starter Story highlights that personal trust and audience building are powerful defenses. One founder notes that while big tech can scrape data, they cannot easily replicate the "know, like, and trust" factor built through consistent YouTube presence, which served as his "insurance policy" when AI overviews threatened his SEO traffic Starter Story — I Make $250K/Month From 13 Businesses (After Losing Everything to AI) @ 03:04.
* Deep Workflows: Sustainability comes from solving the "hundred small problems" specific to a niche, such as complex legal or healthcare compliance, which general models like ChatGPT are unlikely to prioritize My First Million — The Simple Way to Create More Luck, Friends, and Opportunity @ 18:17.

Where they disagree

There is a slight tension regarding the viability of "pointy feature" apps. Starter Story advocates for starting with "micro-SaaS" or single-feature apps that solve one specific pain point deeply Starter Story — How I Built a $12K/Month Micro-SaaS @ 15:15. Conversely, Greg Isenberg warns that the pricing for such commodity tools will likely "ultimately go to zero" because hundreds of competitors can "vibe code" a replica in a weekend Greg Isenberg — If I wanted to build $1M+ AI startup in 2025, I'd do this @ 00:00.

— Sources: 14 videos across 4 creators