Future of Models and Market Consolidation

Agnes AI: Southeast Asia's Answer to ChatGPT (And 20x Cheaper) - Bruce Yang

Topic 12 of 16

Published Feb 2, 2026 · Clip 43:28–47:48 (4:20)

Bruce predicts most models won't exist in three to five years because there is no huge dominance and the landscape is disruptive. He argues the model itself is not a durable moat—ChatGPT was number one a year ago but Gemini 3 is now considered better—so winning products merge good models with a high-usage product. Features, use cases, and product experience matter more to users than the underlying model. He expects a few dominant models serving few products, with smaller models relegated to the long tail.