Table 2. Comparison of traditional vs. generative AI-based quality prediction in meat industry

Criteria	Traditional methods	Generative AI methods	References
Prediction accuracy	Moderate (rule-based or fixed statistical models)	High (learns complex patterns autonomously)	Abuhani et al. (2025) Sarker et al. (2024)
Adaptability to new data	Limited (manual updates required)	High (supports continuous learning and fine-tuning)	Song et al. (2025) Qi et al. (2023)
Data requirements	Requires structured, labeled datasets	Can utilize unlabeled or synthetic data for training	Guo and Chen (2024)
Real-time analysis	Low to moderate (depends on algorithm complexity)	High (enabled by optimized architectures and edge computing)	McCall (2025) Wang et al. (2025)
Explainability	High (interpretable regression/classification models)	Medium (varies with model type; ongoing XAI research)	Arrighi et al. (2025)
Simulation & augmentation	Not supported	Strong support through synthetic data generation and scenario modeling	Fu et al. (2023) Zhang et al. (2024a)