Artificial intelligence is comprehensively transforming the food safety governance system by integrating modern technologies and building intelligent control systems that provide rapid solutions for the entire food supply chain from farm to fork. This article systematically reviews the core applications of AI in the orbit of food safety. First, in the production and quality control of primary food sources, the integration of spectral data with AI efficiently identifies pest and disease, food spoilage, and pesticide and veterinary drug residues. Secondly, during food processing, sensors combined with machine learning algorithms are utilized to ensure regulatory compliance and monitor production parameters. AI also works together with blockchain to build an immutable and end-point traceability system. Furthermore, multi-source data fusion can provide personalized nutrition and dietary recommendations. The integration of AI technologies with traditional food detection methods has significantly improved the accuracy and sensitivity of food analytical methods. Finally, in the future, to address the increasing food safety issues, Food Industry 4.0 will expand the application of AI with lightweight edge computing, multi-modal large models, and global data sharing to create a more intelligent, adaptive and flexible food safety system.

The detection of pesticide residues in food is crucial for ensuring food safety, safeguarding public health, and promoting sustainable development. Overusing pesticides on agricultural crops can lead to the emergence of various diseases. Traditional methods for detecting pesticides offer high precision with limitations like high cost, the requirement of expert technicians, and tedious analytical procedures. To address these issues, nanozymes have been widely applied due to their advantages such as low cost, high stability, and high sensitivity. This review summarizes the research progress of nanozymes in the detection of pesticide residues in food over the last decade, focusing on the synthesis strategies and catalytic mechanisms of carbon-based, metal-based, metal-oxide-based, metal–organic framework (MOF)-based, fluorescence-based, and other X-based nanozymes. This review covers the application of multimodal sensing based on nanozymes in the detection of pesticides, including colorimetric/fluorescence, fluorescence/photothermal, photothermal/colorimetric, and other multimodal sensing techniques. Finally, this review discusses the main challenges currently faced by nanozymes in the detection of pesticides and the current applications of using AI with nanozymes. It also presents future development prospects, with the aim of providing references for the selection of X-based nanozymes and the choice of appropriate detection methods when dealing with traditional and new pesticides in combination with AI.

Food safety requires point-of-care testing (POCT) for mycotoxins, since their presence in wine significantly impacts the wine industry and poses a severe threat to human life. Traditional detection methods are usually limited to detecting one mycotoxin and cannot achieve high-throughput, automated, and rapid quantitative analysis of multiple mycotoxins in real samples. Here, we propose a portable automated microfluidic platform (PAMP) integrating a chemiluminescence (CL) imaging system and a microfluidic chip to realize POCT for multiple mycotoxins in real samples, simplifying complex manual operations, shortening the detection time, and improving the detection sensitivity. Specially, silicone films were used as substrates on microfluidic chips to incubate mycotoxin conjugations, and the streptavidin–biotin (SA-B) system and an indirect immunoassay were implemented on silicone films to improve the sensitivity of reaction results. Interestingly, these methods significantly improved detection results, resulting in sensitive detection of mycotoxins, including zearalenone (ZEA) ranging from 1 to 32 ng/mL, aflatoxin B1 (AFB1) ranging from 0.2 to 6.4 ng/mL, and ochratoxin A (OTA) ranging from 2 to 64 ng/mL. The recovery of samples reached 91.39–109.14%, which verified the reliability and practicability of the PAMP. This PAMP enables sensitive and rapid detection of multiple mycotoxins in markets or wineries that lack advanced laboratory facilities. Therefore, it is essential to develop a portable microfluidic platform for POCT to detect mycotoxins in real samples.

The assurance of food safety requires sensitive monitoring of multiple mycotoxins due to their severe impacts on the food industry and high health risks posed to consumers. Herein, we proposed a chemiluminescent/colorimetric dual-signal readout microfluidic method, incorporating a streptavidin-biotin-alkaline phosphatase (SA-Biotin-ALP) signal amplification system for the highly sensitive detection of Deoxynivalenol (DON), Ochratoxin A (OTA), and Aflatoxin B1 (AFB1). The indirect competitive enzyme-linked immunoassay (ic-ELISA) was integrated into microfluidic chip, resulting in sensitive detection ranges of DON in the range of 4–128 ng/mL, 2–64 ng/mL for OTA, and 0.2–6.4 ng/mL for AFB1, with the limit of detection (LOD) being 2.636 ng/mL, 1.492 ng/mL, and 0.131 ng/mL, respectively. Recovery rates in beer samples ranged from 91.93% to 109.31%. Furthermore, a dual-mode microfluidic workstation (DMMW) was developed to facilitate rapid, automated detection for these mycotoxins, simplifying the detection procedure, enhancing the detection efficiency, and reducing the requirement for specialized personnel, thus confirming significant potential for the rapid detection of mycotoxins in complex matrices such as beer.

Food safety requires real-time monitoring of mycotoxins in food, as food products contaminated with these toxins poses major threat to human health. In this study, we proposed a remote-controlled microfluidic platform (RCMP) integrated with chemiluminescent/colorimetric detection system for rapid, cost-effective and real-monitoring of multiple mycotoxins in real samples based on the indirect competitive enzyme-linked immunosorbent assay (ic-ELISA). The RCMP enabled sensitive and automatic detection of deoxynivalenol (DON), zearalenone (ZEA), and fumonisin B1 (FB1) in the range of 4–128 ng/mL, 1–32 ng/mL, and 0.5–16 ng/mL, respectively. The limits of detection (LOD) were 2.881 ng/mL for DON, 0.702 ng/mL for ZEA, and 0.470 ng/mL for FB1. In further validation, satisfactory recoveries between 93.57% to 108.47% with the relative standard deviations (RSDs) of 6.92–11.39% were obtained in beer samples. Overall, RCMP provides an automatic, high-throughput and cost-effective method for detection of DON, ZEA, and FB1 and can be confidently applied for monitoring in beer samples.