While primarily considered a respiratory disease, COVID-19 can result in multi-organ pathologies, including autoimmune and autoimmune-like complications. With other viral infections known to be associated with, or exacerbate autoimmune diseases through molecular mimicry, we hypothesized it could also contribute to COVID-19 associated complications. Indeed, regions of sequence identity and similarity between SARS-CoV-2 and human proteins have been found in silico, albeit no studies have validated these in clinical samples. Previously, we applied novel immune cross-reactivity criteria and identified 11 novel predicted B cell-epitopes within human proteins, with the potential to be cross-reactive with SARS-CoV-2 and explain aspects of COVID-19 pathology (1). Herein, plasma or sera collected from cohorts of both COVID-19 negative and convalescent volunteers, from Victoria and Tasmania (n=290), was tested by indirect ELISAs to measure IgG specific antibodies to the full-length SARS-CoV-2 epitopes and their corresponding epitopes identified in human proteins. Responses were confirmed towards 7 SARS-CoV-2 specific epitopes and 6 corresponding epitopes in human proteins. Two of the novel SARS-CoV-2 epitopes from the Nucleoprotein (NP) were recognised in most of the cohort, regardless of COVID-19 status (>83% positive), indicating potential environmental cross-reactivity. Overall, there was substantial donor heterogeneity, with most individuals responding to only either SARS-CoV-2 or it’s corresponding potentially cross-reactive human protein derived epitope, plus a small number of individuals who reacted to both. Overall, our study has identified novel B cell epitopes within human proteins, which mostly retain specific narrow recognition, and exhibit cross-reactivity potential in only a small subset of individuals. This unexpected finding highlights the need to validate predicted epitope cross-reactivities, even in largely identical sequences, testing in sufficiently comprehensive large clinical cohorts.