Motif AI automatically discovers hidden web workflows to automate

Motif tackles a key limitation of current LLM-based automation tools: they assume users already know what to automate. Instead, Motif silently observes everyday browsing—clicks, form fills, page navigations—and uses a pattern-discovery algorithm to spot sequences that recur and could be turned into programs. When a pattern is found, Motif suggests an automation to the user. If confirmed, it generates a runnable script that the user can review and tweak using natural language commands, making it accessible even to non-programmers.

In a multi-day study with eight participants, Motif uncovered more automatable workflows than participants could come up with via "vibe coding" (ad-hoc prompting). Most discovered patterns aligned with users' actual routines and were deemed useful. Follow-up surveys indicated strong intent to continue using Motif-generated programs. The system represents a shift from reactive automation (users explicitly ask for a script) to proactive discovery, potentially making web task automation far more widespread.