Integrating human and nonhuman intelligence to reduce tobacco's public-health burden

Tobacco use remains the leading preventable cause of death worldwide, while the rise of electronic nicotine products has sparked a new wave of initiation. The urgent need for scalable, multilevel tobacco-control interventions converges with the rapid advances in artificial intelligence (AI). This article reviews the most recent literature on integrating machine- and human expertise to enhance tobacco-cessation strategies within a multilevel framework.

Recent advances in predictive analytics, large-language models (LLM), AI chatbots, and related tools create a framework to strengthen tobacco prevention. Predictive analytics merge electronic health records, behavioral surveys, genetics, and real-time sensor data to model the complex multilevel factors that influence quitting. LLMs instantly uncover informative features, revealing novel predictors that shape targeted interventions. AI-driven conversational agents deliver stage-specific counseling and medication guidance, with preliminary trials showing improved engagement and quit rates. Reinforcement learning personalizes messaging, rewards, and medication schedules to optimize outcomes, while natural-language processing of social media provides fine-grained sentiment data to assess policy impact.

Realizing AI's potential to reduce tobacco's public-health burden requires interdisciplinary collaboration, equity-oriented design, external validation, and strong governance. These safeguards can enable scalable, adaptable, and culturally relevant smoking-cessation interventions and facilitate timely, effective tobacco-control policies.