In the heart of Kampala, Uganda, a fascinating investigation delved into the intricacies of how TikTok's algorithm shapes the way news is shared and consumed among residents of Kibuli Parish. The core aim of this research was to unravel the complex dynamics of how the platform's algorithmic framework affects the visibility, variety, and consumption habits of news-related content, with a specific focus on the emergence of information silos. By adopting a multimodal research strategy, the study seamlessly integrated the quantitative examination of TikTok videos with in-depth, qualitative conversations with local TikTok enthusiasts. A carefully curated group of 40 active news consumers on the platform was selected using a targeted sampling approach, providing a rich tapestry of insights. The data collection process involved the use of rigorously designed questionnaires, interview and content analysis, which were subsequently analysed through a combination of statistical summaries and thematic deconstruction. The results revealed a dual-edged phenomenon: while TikTok's algorithm dramatically boosts the visibility and engagement of news stories, it also inadvertently creates pockets of like-minded individuals and reinforces pre-existing biases, thereby limiting the diversity of perspectives. Furthermore, the algorithm's propensity to prioritize sensational content facilitates the rapid dissemination of misinformation, often overshadowing locally relevant stories. Ultimately, the study concludes that although TikTok has revolutionized news dissemination in Kibuli, its underlying algorithmic architecture poses significant challenges, including diminished content diversity and heightened polarization. To address these concerns, the researchers propose a multifaceted approach, encompassing enhanced content regulation, the promotion of diverse and inclusive content, and increased transparency in algorithmic decision-making, with the ultimate goal of fostering a more responsible and nuanced news consumption culture among users.