e21017 Background: Access to cancer information on the Internet has proliferated, yet ensuring the accuracy and reliability of such information remains a challenge. The abundance of online content necessitates a comprehensive understanding of the prevailing landscape of cancer information dissemination. Methods: Cancer-related information was retrieved from search engine results pages of NAVER and Google, the most frequently used websites for information search in South Korea, during September and October 2023. Promoted posts and overt advertisements were excluded from the dataset. Posts featuring more than 50% of advertising content were categorized as fully hidden advertisements (HA), while those with lesser proportions were classified as partially HA. In cases where there was no advertising content but the hospital name or contact information was indicated for promotional purposes, it was classified as a simple HA. Topic modeling techniques were employed to analyze the content of HA posts, identifying prevalent themes and keywords indicative of promotional intent. Results: Of the collected posts, 48.6% were classified as HA. The publisher of the HA was 26.8% of oriental medicine/care hospitals, followed by general hospitals (25.3%), non-professional (21.3%), and small and medium-sized private hospitals (17.4%). In HA, 40.3% were full advertisements, 36.75% were partial advertisements, and the remainder were simple displays. By cancer type, 65.3% of posts about breast cancer were advertising posts, 55.2% were about colon cancer, and 53.7% were about stomach cancer. In the topic modeling analysis of HA, despite containing basic and theoretical information about cancer, HA subtly directs users towards advertising content, featuring terms related to private insurance and complementary medicine. Conclusions: The study elucidates the prevalence and characteristics of HA within cancer-related information available online, highlighting the need for heightened awareness. While it may not be inherently wrong, HA can impact user perception and decision-making processes. Additional research is underway to determine whether information classified as HA can be trusted. Addressing issues such as HA within cancer information will likely require cooperation between search engine platforms and government agencies.