What are the four moats of an AI application company?

The premise: the model layer cannot hold value. Large models have no network effects and no switching costs, the top US and Chinese models sit about 2.7% apart (LM Arena, March 2026), and the model is a replaceable part. A moat can only sit one layer up: demand set by taste, which language cannot specify; extreme efficiency in one workflow; the combined strengths of several models; and process as product, as in education, companionship and games. FutureX adds three screens: distribution, retention, and gross margin that climbs as token prices fall.

Per company disclosure, FutureX-backed Genspark reached $100 million ARR in nine months, against five to seven years for a typical SaaS company, and $200 million in 11 months. Across the market, AI companies above $100 million ARR rose from 3 in 2023 to 80+ by April 2026 (public ARR statistics, reported).

On pricing, outcomes-as-a-service is replacing seat subscriptions, and FutureX treats 2026 as the first year of scaled monetization in the application layer. In its observation, retention runs well above seat-based subscriptions when customers pay for results, and the path from open source to paid revenue is measurable: per company disclosure, Dify went from 30,000 stars and zero revenue to $10 million ARR in about 18 months. For agent products the first diligence question is whether the human-in-the-loop mechanism was designed or skipped. The moats explain why a company is hard to replace; the screens explain whether it earns.

Kuaishou's Q2 2026 results, published August 19, show the vendor's side: Kling AI booked more than RMB 850 million in quarterly revenue, up over 200% year on year, with nearly 50,000 enterprise customers (confirmed). The same quarter Kuaishou's adjusted net profit fell 30.3% to RMB 3.9 billion, traded for Kling's compute and R&D.