The S18 cohort focused on Applied AI, moving beyond research into vertical-specific predictive models for consumer behavior and creative workflows. Founders targeted the D2C and enterprise marketing sectors, leveraging early machine learning to automate high-intent decision-making. This batch was notable for proving that AI could move from a "nice-to-have" feature to a core recommendation and forecasting engine.
These startups tapped into the massive demand for hyper-personalization and market foresight in crowded consumer categories like wellness and beauty. By turning fragmented search data and health profiles into actionable insights, they proved that users were willing to pay for algorithmic curation over manual discovery. They successfully commercialized the shift from static data to dynamic, predictive intelligence.
Many S18 AI startups struggled because their models relied on expensive, proprietary data pipelines and lacked the generative flexibility of modern LLMs. The "black box" nature of early predictive tools often led to trust gaps when forecasts didn't align with rapid market shifts. Builders today must realize that prediction without generation often leaves the user with a problem but no immediate solution.
A solo founder can now rebuild the automated trend-to-product pipeline by combining predictive analytics with real-time generative design. Instead of just forecasting a beauty trend, use multi-modal LLMs to instantly generate product formulations and marketing creative. The wedge is no longer just "knowing what's next," but automating the execution of that insight.
Personalized daily vitamin packs and dietary guidance powered by AI for Chinese consumers.
AI-powered trend forecasting platform that predicts consumer trends across beauty, wellness, and food industries with 72% accuracy.