All models have a bias. Bias occurs because models are a mathematical representation of a highly complex system which is a simplification of the reality of many processes. Models can have bias because of limiting spatial resolution or incomplete knowledge of how a process works.
For example, some climate models estimate the timing of the monsoon arrival to be earlier than expected in reality. This can lead to outputs of temperatures being too high or too low and can give unrealistic results. Model bias is important to consider when choosing or using climate projections.
Briefing: Climate modelling – an overview
Briefing: A closer look at climate modelling