What is the role of AI in forestry?

AI's role in forestry is to read more data than people can and turn it into something a person can check and act on. In forest monitoring that means satellite, LiDAR and field data turned into maps of cover, volume, carbon or disturbance. In the business of forestry it means contracts, scale records, tickets and invoices turned into answers about volume, cost and money owed. In both, the output is only as good as what it is checked against: sample plots and inventory data for monitoring, and the business's own trusted records for operations.

What the usual answer says

The top results for this question list applications: monitoring biodiversity, detecting fire and pests, measuring timber from photographs, estimating carbon, planning logistics. Each is real. A list, though, cannot say which applications a forestry business should trust, or why some work in practice and others stay in pilots. The research base behind those lists is young. A 2026 review in Forestry by Subedi, Gautam and LeBel found 69 documents on AI in the forest products supply chain, 55 of them original research. The literature shows "massive growth from 2019 onwards", with the highest number in 2024.

What the FAO says the role is

The FAO's forest monitoring team set out AI's role in a February 2026 blog post by Pilar Valbuena and colleagues. They name three jobs: "Automating labour-intensive analysis and reporting tasks", building "User-friendly and interactive decision-support systems", and "Analysing complex and diverse datasets that could not easily be integrated" by conventional methods.

The team attaches a condition. For outputs that must stand up to policy scrutiny, it writes, "remote sensing outputs need systematic calibration and independent validation". It also summarises a 2026 review in Remote Sensing of more than two decades of LiDAR-based vegetation monitoring. That review found LiDAR- and AI-derived forest metrics most reliable when anchored in field protocols and national inventory data. It named "persistent limitations related to cross-sensor calibration, data harmonisation, and interoperability". In monitoring, the check is statistical: sample plots against a map, not every pixel against a record.

What practitioners worry about

Researchers at the NAU School of Forestry interviewed 20 forestry professionals across academia, government and industry, for a study in Forest Policy and Economics reported in October 2025. The professionals saw AI as useful for routine paperwork and summaries. They also saw great potential in it for complex analysis. The lead researcher, Alark Saxena, told the university's news office that this held "as long as it functions as an assistant that enhances, rather than replaces, the judgment of an experienced professional". Their main concerns were the black box, AI whose reasoning nobody can follow, and "the risk of training AI using some agencies' poor-quality or biased data and then trusting its flawed outputs for important land management or policy analysis."

Both concerns are about the same thing as the FAO's condition: an output nobody can check.

The same condition on the business side

On the business side, the equivalent of the field plot is a record kept by someone other than the system being checked. AI's role there is to read many records together, faster and more completely than a person can, and to flag where they disagree. Unlike monitoring, the check can often be made record by record, because the documents exist for every load and every invoice. Our argument is that this is where AI earns its place in the office. Summarising one system's figures adds little; setting them against another's is the job.

How to use the condition

Before handing AI a forestry task, name the record its output will be checked against. For a stand volume estimate, it is the cruise or the scale. For a forecast, it is what was actually harvested. For an extracted contract rate, it is the invoice that applied it. If there is no such record, the task is research, and the output should be treated as a hypothesis.

When it doesn't apply

Exploratory research uses AI to find patterns nobody has measured yet, and by definition there is no record to check against. Much of monitoring, such as change detection over large areas or early pest detection, exists because no affordable record could be made in time. There the check is a sample, taken after the fact, and some tasks, such as spotting a fire, are acted on before any check is possible. Those cases set their own tolerance for error.

Quarri for forest management is built around how a forest operation runs, from the cruise to the settled account.

Quarri is an AI-native data platform for the timber supply chain. It connects buying, production, sales and inventory for forest management, sawmill, wood products and pulp, paper and packaging operators.