Loneliness is associated with mental health and thus is of particular concern in the COVID-19 pandemic, due to physical distancing restrictions and shelter-in-place orders. The current study assessed the associations of age, gender and their interaction with loneliness during the COVID-19 pandemic, controlling for other sociodemographic variables. A pooled sample of 3,012 English-speaking Canadian adults aged 18+ years completed a web-based survey in one of three waves between May 8 and June 23, 2020. Multivariable logistic regression was used to examine the associations of loneliness with age and gender controlling for marital status, household income, education, living alone, employment situation, and survey wave. A likelihood ratio test assessed the model with interaction between age and gender included. Approximately 8.4% of the sample reported feeling lonely 5+ days in the past week. The regression model with main effects found greater odds of loneliness among women than men (AOR = 1.76, 95%CI = 1.32, 2.34) and among all age groups younger than 60 years compared to those aged 60+ years (p = 0.002). In the final regression model, a significant interaction effect between age and gender on loneliness was found. The interaction showed that women had greater odds of loneliness than men among those aged 18-29 years (AOR = 3.53, 95%CI = 1.69, 7.37) and 60+ years (AOR = 2.62, 95%CI = 1.33, 5.17). Special consideration of loneliness among younger and older adult women is needed in service planning. Given inconsistencies with pre-pandemic studies, detailed data collected during the current crisis is essential to inform proactive resource allocation to prevent and treat mental health consequences of the pandemic.