Electricity load forecasting algorithms are used in the smart grid to predict the electricity demand in the future. Besides that, it helps in reducing the electricity generation cost. In the literature, three types of load forecasting are done such as short-term, medium-term, and long-term. In this paper, the short term forecasting is done. The short-term forecasting algorithm predicts the electricity demand from a few hours to several weeks ahead. Due to the nonlinear, nonstationary, and non-seasonal nature of the electric load time series, accurate forecasting is challenging. In this paper, Artificial Intelligence (AI) and the swarm optimization algorithm is hybrid in order to improve the prediction of load forecasting. We have considered Artificial Neural Network (ANN) and Binary Particle Swarm Optimization (BPSO) algorithms in our work. The BPSO algorithm used to improve the learning rate in the ANN network. The experimental results were simulated in MATLAB and various performance metrics such as RMSE, MAPE, minimum and maximum error determined. The results show that the proposed algorithms provide better results as compared to the existing algorithms.