Simulating trustworthy content delivery on social media by manipulating algorithms and architecture

This research investigates the trustworthiness of social media platforms in providing access to reliable information. With millions relying on social media for timely and accurate information, poor quality information significantly erodes trust. We examine the interplay between two key elements of social media platforms that are responsible for content delivery: the algorithms that dictate content visibility and the architectural frameworks that govern how platform components and users relate to each other, which in turn shapes how users interact and the content that they see from other users. We develop a simulation model that evaluates how different classes of social media platforms (as defined by their algorithms and architectures) handle the spread of information of varying levels of quality. We find that TikTok shows the fastest and widest information spread, driven by its highly connected architecture, while Twitter, Reddit, and Facebook diffuse more slowly. Moreover, algorithmic design further shapes outcomes, with engagement-driven ranking amplifying virality and producing long-tail popularity effects.