Background: Pulmonary vein isolation (PVI) has become the cornerstone of atrial fibrillation (AF) ablation, but long-term success rates remains suboptimal, due in large part to late PV reconnection and insufficient ostial substrate modification. Objective: To evaluate whether ablation index (AI)–guided PVI with electrical isolation and quantified ostial substrate modification improves clinical outcomes when compared with contact force (CF)–guided ablation in patients with paroxysmal atrial fibrillation (PAF). Methods: The PAF-AI trial (ChiCTR1900022041) is a prospective, multi-center, randomized controlled clinical trial enrolling patients with PAF with an indication for catheter ablation. Patients are randomized into a 2:1 fashion to two treatment arms: AI-guided PVI (n = 151) and CF-guided PVI (n = 75). In the AI-guided PVI group, real-time automated display of radiofrequency applications (Visitag™) is used with AI ≥ 500 recommended at the anterior/superior/inferior walls and 350–400 at the posterior wall. In CF-guided PVI group, the value and direction of CF are displayed, with the lesion dots manually annotated. The primary endpoint is the freedom from AF recurrence at 12 months following ablation, without antiarrhythmic drug. The primary pre-specified secondary endpoints include intraprocedural efficiency and peri-procedural complications. Conclusions: PAF-AI trial compares the effectiveness and safety of two different strategies of PVI in patients with PAF, AI-guided PVI versus more established CF-guided PVI. This prospective, multi-center, randomized controlled trial, with comparative data evaluating procedural and long-term follow-up results, aims to evaluate the impact of AI-guided strategy on AF ablation compared with the current standard of care RF ablation approach. © 2020, Springer Science+Business Media, LLC, part of Springer Nature.