Generative AI in a new era of computer model-informed tobacco research: a short report

Tobacco Control, Journal Year: 2025, Volume and Issue: unknown, P. tc - 058887

Published: Jan. 30, 2025

Background Social media influencers who promote e-cigarettes on Instagram or TikTok for tobacco brands use marketing tactics to increase the appeal of their promotional content, example, depicting alongside healthy lifestyle entertainment imagery that could decrease youths’ risk perceptions e-cigarettes. Monitoring prevalence such content social using computer vision and generative AI (artificial intelligence) can provide valuable data regulatory science (TRS). Methods We selected 102 videos posted by micro-influencers in 2021–2024 promoted posts featuring four themes: cannabis, entertainment, fashion lifestyle. used OpenAI’s GPT-4o multimodal large-scale visual linguistic model detect presence nicotine vaping, cannabis fashion, The did not require any additional training improved its performance as we modified text prompt. Results model’s accuracy was 87% 96% 99% 98% Conclusions Generative achieve accurate object detection with zero-shot learning (no pretrained model). This be applied big data-scale sample sizes images e-cigarette-related other substance-related contexts (eg, lifestyle) promotion these products media, providing TRS.

Language: Английский

Public Perceptions and Discussions of the FDA JUUL Ban Policy on Twitter: Observational Study (Preprint)

JMIR Formative Research, Journal Year: 2024, Volume and Issue: 8, P. e51327 - e51327

Published: May 15, 2024

On June 23, 2022, the US Food and Drug Administration announced a JUUL ban policy, to all vaping electronic cigarette products sold by Juul Labs.

Language: Английский