The human learning process relies on cognitive empathy, developed from early childhood through observation and personal experience. In artificial systems, empathy predicts an uncertain future based on observed and subjectively perceived environmental states and other agents. Learning from others involves an agent projecting itself into the future using an imprecise representation of others’ knowledge, derived from subjective interpretation, data aggregation, and interpolation based on its own experience. Artificial empathy enhances learning, adaptation, and decision-making in cooperative scenarios by integrating others’ knowledge, visualizing their states, and predicting potential actions and outcomes when mimicking them. A model of imprecise cognition–using the agent’s knowledge, imprecise representations of others’ states, and environmental inputs–supports future state prediction under uncertainty and selecting actions that maximize goals. The model’s performance is demonstrated in a robot swarm, where empathetic actions and successful collaboration yield clear rewards.