Explainable AI Comes to Vertical Farming

Research advances suggest transparent AI models could improve cost control and oversight in indoor agriculture

13 Feb 2026

Vertical farming groups are exploring a new generation of artificial intelligence systems designed not only to automate crop production but also to explain their decisions.

Indoor farms have long relied on AI to regulate lighting, temperature, humidity and carbon dioxide levels. These systems process continuous streams of sensor data to optimise plant growth. In most commercial operations, however, the underlying models operate as “black boxes”: growers can see the results, but not the logic behind each adjustment.

Explainable AI seeks to address that gap.

In research settings, particularly in greenhouse-based controlled environment agriculture, explainable models have shown strong forecasting performance while identifying which environmental factors drive specific outcomes. Some pilot studies report high accuracy rates. Yet these trials remain largely confined to academic projects rather than full-scale commercial vertical farms.

Artificial intelligence is already embedded in many large operators. Companies such as AeroFarms and Bowery Farming use proprietary platforms to interpret sensor data, track crop health and fine-tune growing conditions. Public evidence of fully explainable systems operating at scale, however, remains limited. Commercial tools tend to prioritise automation and yield performance over transparent decision-making.

The potential commercial case is nonetheless attracting interest. Energy is among the largest costs in indoor agriculture, where lighting and climate control operate around the clock. A system that not only recommends adjustments but also clarifies why they are required could help operators refine energy use, reduce waste and strengthen internal oversight. Greater transparency may also support compliance as sustainability and food safety standards tighten.

The industry’s financial backdrop adds urgency. After a period of rapid expansion followed by funding pressures and consolidation, vertical farming groups are under pressure to demonstrate operational discipline. Investors are increasingly focused on resilient, data-driven models that balance innovation with cost control.

Adoption of explainable AI remains at an early stage. Reliable sensor networks, robust data infrastructure and careful system integration will be necessary before widespread deployment is feasible.

As controlled environment agriculture matures, the competitive advantage may shift from automation alone to systems that can account for each decision they make.