Global warming increases the frequency with which drought and heat stress occur simultaneously, especially in semi-arid regions. Such combined stress imposes a non-additive and more severe impact on plant growth, yield, and quality than either stress alone. Here, we integrate recent physiological, biochemical, and multi-omics studies to compare individual and combined stress responses and to dissect the underlying signal transduction networks. We show that drought-dominated phases rapidly elevate ABA concentrations and activate SnRK2–AREB cascades, whereas heat pulses trigger jasmonic acid and ethylene signals that antagonize ABA-driven stomatal closure. Under combined stress, these hormonal modules converge on a “competitive TF marketplace”, where ABA, JA, and GA cis-elements co-regulate invertase–sugar checkpoints, heat shock factor/ROS oscillators, and chromatin-remodeling events that determine reproductive fate. Recent advances using multi-omics approaches and systems biology have further elucidated these complex networks. These insights will inform future breeding strategies aiming to develop stress-tolerant crops. We highlight emerging tools—weighted gene co-expression networks, kinetic multi-omics, and cis-regulatory CRISPR editing—that can exploit these signaling hubs for breeding crops with improved combined stress tolerance.

With the growing global population, intensifying resource constraints, and deepening climate change impacts, agriculture faces dual challenges of ensuring food security and advancing sustainable development. Artificial intelligence (AI) has emerged as a transformative technology, penetrating the entire crop production chain and offering innovative solutions to traditional agricultural bottlenecks. This paper systematically reviews AI applications in five core domains: biotic stress monitoring, soil health management, precision operation, supply chain optimization, and climate-resilient agriculture. It further categorizes and analyzes four key technical pathways—deep learning, sensor fusion, data-driven methods, and hybrid modeling—while critically examining major challenges across data, technology, implementation, and ethics/policy dimensions. Future directions are discussed from technological innovation, scenario expansion, implementation guarantees, and sustainability orientation. Research findings show that AI has achieved technical validation in pest/disease detection, soil parameter modeling, and intelligent spraying, with accuracy exceeding 85% in some cases. However, regional data bias, insufficient model generalization, and the digital divide still hinder large-scale deployment. Moving forward, coordinated efforts in technological innovation and policy support are required to promote inclusive, standardized, and sustainable AI applications in crop production.

Combined drought–heat stress is not adequately described by adding the responses to drought and heat in isolation. Water limitation restricts evaporative cooling, whereas high temperature increases atmospheric demand and threatens photosynthesis, proteostasis, and reproduction. To assess what has been demonstrated rather than merely proposed, we combined a structured narrative review with a study-level evidence map spanning genomics, multi-omics, and emerging artificial intelligence (AI). The PubMed search, updated on 9 August 2026, returned 254 records; targeted and backward searches contributed eight seminal or reviewer-nominated publications. Independent screening retained 90 verified publications: 69 direct combined-stress studies, 11 genomics studies, six original AI or prediction studies, and four reviews or quantitative syntheses used for context. No conserved molecular signature emerged. Response direction varied with soil water status, vapor-pressure deficit, stress order, tissue, and developmental stage. Carbon allocation, redox and proteostasis balance, reproductive failure, and recovery recurred across omics layers, although neither their direction nor their adaptive value was consistent. Genomic loci and predictive performance were also environment dependent. Direct uses of AI remain uncommon and currently support prioritization more convincingly than causal inference. We organize this uneven evidence as a ladder extending from molecular association to functional validation, multi-environment prediction, and cultivar deployment. The main shortage is not another nominal omics layer but harmonized factorial experiments that measure recovery and reproduction, report field-relevant environmental metadata, and include external validation.

Late sowing has been common since the popularization of mechanical transplantation, and it has led to reduced grain yield due to low grain weight. However, the uneven contribution of the amount and weight of grains in different positions to grain yield remains unclear. A 2 yr field experiment was conducted with two super rice varieties (Ningjing 7th: N7 and Nanjing 9108th: N9108) and three sowing dates (S1: 30 April; S2: 30 May; and S3: 30 June) in Hefei City. In this study, delaying sowing shortened the growth duration, reduced temperature, and further reduced the grain yield mainly by suppressing the total grain weight. Compared with S1, the grain weight of grains in secondary branches (SGs) in S2 and S3 was reduced by 11.1% ± 2.9% and 18.5% ± 1.4%, respectively, due to a lower reproductive-stage temperature. However, no significant difference was found in the grains in primary branches (PGs). Moreover, a shorter growth duration enhanced the ratio of tillers in a lower position (LT) per unit area, which contained more SGs per panicle, and finally led to a higher ratio of SG amount per unit area (SG%). The SG% increased by 13.4% ± 1.4% and 21.1% ± 1.9% in S2 and S3 compared with S1, respectively. In conclusion, delaying the sowing date mainly reduced the grain weight of SGs and enhanced the SG%, leading to a lower apparent grain weight and further decreasing grain yield.