The high production of information during the COVID-19 pandemic, characterized by the WHO as an “infodemic”, has fueled uncertainties and anti-science discourse, particularly on social media. This phenomenon has significantly contributed to vaccine hesitancy, evident in Brazil’s low childhood vaccination rates against COVID-19, where only 10% were fully vaccinated. To address this challenge, this study analyzed over 1.8 million messages shared in 248 Telegram anti-vaccine channels between July 1, 2022 and January 31, 2023. Using a methodological approach based on the use of data science and data analysis techniques, we identified five common types of vaccine misinformation: 1) pseudo-protective (propagating pseudo-protective measures against ‘experimental vaccines’); 2) immunological (building clinical stories that would demonstrate the reduction of immunity due to immunization); 3) false utility (disseminating legal and argumentative instructions for parents not to vaccinate their children); 4) medicalizing (highlighting medical reports to legitimize false or inaccurate facts); and 5) conspiracist (spreading conspiracy theories based on the idea that global corporations want to subjugate people with eugenic policies). These findings offer valuable insights for health professionals and managers to strategically develop targeted communication campaigns to counter these narratives and address vaccine hesitancy.