The socio-cultural polarization induced due to information and communication technology because of the selective online exposure during COVID-19 has been a major cause of concern around the globe. In this paper, we use random network theory-based simulation technique to investigate the temporal dynamics of opinion formation on YouTube videos. Our findings reveal that as the pandemic unfolded, the degree of polarization in the online discourse has increased with time. This study is significant for understanding that online discourse on sociocultural issues can lead to polarization particularly in crisis situations such as a pandemic and exacerbate the social divide.