The development of artificial intelligence should avoid the harm caused to the public by algorithm design, and must clarify the algorithm motive and interpretability to overcome the unfair influence caused by algorithm design and data collection.