Food safety is crucial to global public health. In recent years， new hazards have arisen from changing consumption patterns and novel food sources， alongside traditional threats like foodborne pathogens， mycotoxins， and chemical residues. The rapid advancement of artificial intelligence (AI) technology provides innovative solutions to address these challenges. This article summarizes the applications of AI technology in the proactive prevention and control of food safety， including risk warning， toxicity prediction， rapid detection， and efficient prevention and control strategies. AI integrates meteorological statistics， mechanistic models， and machine learning algorithms to achieve early warning of risk factors such as foodborne pathogens， mycotoxins， pesticides， and heavy metals in food. AI also assists traditional toxicological models and incorporates transfer learning to facilitate toxicity prediction and risk assessment of new hazards. Besides rapid detection for food safety and quality， AI also plays a role in high-throughput design and screening of both traditional and novel recognition elements such as antibodies and aptamers. In the realm of proactive prevention and control， AI combines with bioinformatics， molecular biology， and synthetic biology to effectively predict and screen antimicrobial peptides， degrading enzymes， and bacteriophages， revealing their antimicrobial and degradation mechanisms. However， the application of AI in food safety still faces challenges such as insufficient data sharing， standardization， and handling of multimodal data. With the advancements of technology and evolvement of data sharing mechanisms， AI is expected to play an increasingly important role in ensuring global food safety and addressing complex and evolving food safety issues.