The role social media platforms play in contraception mis- and disinformation is well-known. However, existing studies provide only a temporal snapshot, which quickly becomes outdated due to the fast-evolving nature of online discourse. Additionally, descriptive studies of content are often unable to measure the impact of public messages or rely on limited survey campaigns. Finally, as social media platforms are enormous data sets, existing studies are limited to small samplings followed by manual scoring of collected information.
AI technologies can rapidly process vast amounts of textual and multimedia data, enabling real-time identification and analysis of misinformation trends and their evolution over time. Our research group aims to create new, AI-based research tools through which the community can produce ongoing, live queries of contraception information and misinformation posted on social media and measure the impact of messages through aggregation and assessment of user response. For this project, we propose building a pilot tool for the social media platform Reddit. Reddit offers primarily text-based content with posts ideal for nuanced analysis and the ability to do real-time trending. Partnering with an AI software developer, our goal will be to rapidly 1) build the AI-based research tool 2) validate the tool output to ensure high quality, reproducible data and 3) deploy, iterate and optimize the tool to the feedback of the research community. Ultimately our goal is to enable real-time information about contraception mis- and disinformation online through both assessment of content and measurement of impact through AI-enabled collection, processing, and analysis.