TikTok's algorithm did not treat Democrats and Republicans equally during the 2024 US presidential election. According to a paper published in Nature, its recommendation system showed a Republican-leaning skew in three states. The journal's editors have also published a Research Briefing in the same issue covering the study.

. . .

TikTok is a social media platform where users create, share, and discover short-form videos. Its For You page is a personalized recommendation feed that uses an algorithm to decide what individual users see. Unlike Facebook or X, where you choose what or who to follow, TikTok's computer program does most of the picking for you.

Training bots

Researchers at New York University Abu Dhabi, UAE, created 323 dummy (bot) accounts across the states of New York, Texas, and Georgia. These bots were trained to be either Democrats or Republicans by watching videos aligned with partisan content, while others stayed neutral. Then the team tracked the videos TikTok recommended on these accounts' For You pages over 27 weeks.

The results revealed a clear partisan imbalance, as the team explains in their paper: "Republican-conditioned accounts received 11.5% more party-aligned content compared with Democratic-conditioned accounts across all states."

. . .