In this paper, a novel optimization technique is implemented to explore the effects of considering uncertain load positions. Therefore, the integration of reliability-based design into structural topology optimization, while considering imperfect geometrically nonlinear analysis, is proposed. By comparing the results obtained from perfect and imperfect geometrically and materially nonlinear analyses, this study examines the impact of nonlinearity on probabilistic and deterministic analyses. Concerning probabilistic analysis, the originality of this research lies in its incorporation of the position of the applied load as a stochastic variable. This distinctive approach complements the consideration of other relevant parameters, including volume fraction, material properties, and geometrical imperfections, with the overarching goal of capturing the variability arising from real-world conditions. For the assessment of uncertainties, normal distribution is assumed for all these parameters. Normal distributions are chosen due to their advantages in terms of simplicity, ease of implementation, and computational efficiency. These characteristics are particularly beneficial when dealing with complex optimization algorithms and extensive analyses, as is the case in our research. The proposed algorithm is validated according to the results of benchmark problems. Structural examples like cantilever beam, pinned-shell, and L-shaped beam problems are further explored within the context of imperfect geometrically nonlinear reliability-based topology optimization, with specific regard to the probabilistic aspect of the location of the externally applied loads. Moreover, the results of the suggested approach suggest that the inclusion of a probabilistic design strategy has influenced topology optimization. The reliability index acts as a controlling constraint for the resulting optimized configurations, including the mean stress values associated with the resulting topologies.