Forestry remains one of the most demanding high-stakes sectors in the fight against climate change. Yet, until now, corporate applications for artificial intelligence and specifically generative AI, have been deployed mostly in controlled environments, such as offices. As we push for a net zero world, a new frontier for AI exists in its deployment within natural ecosystems that support our planet.
If we are concerned with making large-scale positive environmental impact, forestry is one of the most significant sectors we need to optimise. Even small improvements in forest silviculture and ecosystem services management can have positive implications for biodiversity and global carbon balance.
Why the field matters
More than one-third of the world’s forests are used for economic purposes, according to data from the Food and Agriculture Organization of the United Nations (FAO) Managing commercial forests effectively and sustainably requires high-performing operations as well adherence to regulatory requirements: in the field, foresters need wide knowledge and understanding of laws, regulations and company guidelines. As greater focus is placed on how forests should be effectively managed, regulations consequently become more complex and diverse across jurisdictions. It is an enormous efficiency to be able to check various information on the spot.
Traditional approaches to forest monitoring and management rely on periodic remote sensing, infrequent ground surveys or manual inspection. Remote sensing technologies like satellite imagery and drones provide detailed, real-time data on forest cover, structure and health. On-the-ground surveys and biomonitoring also complement these remote sensing methods but these techniques can also be slow, costly or imprecise. Modern AI-enhanced systems are beginning to close that gap. For instance, Meta and the World Resources Institute now publish global tree canopy height maps at one-meter resolution, enabling detection of individual trees and vastly improving change detection across landscapes.
From analytics to generative intelligence
Until now, AI in forestry has remained analytical: using machine learning for growth and yield improvement, risk management and other decision support systems for, e.g., planning operations. So far, the use of Generative AI is much less reported. In forest environment, tools that can reason, with which one can interact via language or vision and which can respond to unstructured input, can be especially powerful when it comes to real time decision making and access to information on the spot. For example, personalized forest management, sustainable practices, and reskilling & upskilling are also future areas
Enter Aarnibot
Currently, AI systems for field-based work are more to augment, rather than replace workers. In complex settings like forests or farms, human skill, tacit knowledge and contextual judgment remain indispensable.
One example of such system is Aarnibot, a generative AI companion developed by UPM, a Helsinki-based material solutions and forestry company, that turns wood into renewable products. Aarnibot is built to operate in rugged forest terrain. It provides versatile support for UPM personnel who work in the forest across various tasks, such as silvicultural operations, applying regulations, execution guidance, and the use of information systems. This guidance helps to ensure smooth daily operations and supports discussions with forest owners. Additionally, it supports customer-facing roles by providing information about UPM’s services and, when needed, generates promotional texts that can be directly included in customer emails. The intelligence? Simply providing easy access to wide body of knowledge via multi-modal mobile interface, in environment where laptops are a no-go.
Finnish forests (including trees, soil and peatlands) have historically absorbed between 30-60% of Finland’s total emissions, making them the country’s largest natural carbon sink. Due to their importance, Finland’s national climate and forest strategies aim to increase active forest management to enhance carbon sequestration, alongside other measures like promoting bioenergy with carbon capture.
AI systems like Aarnibot can play a pivotal role in active forest management, through supporting Forest Stewardship Council (FSC) certification, which covers more than 200 million hectares of forest worldwide, and regulation on deforestation-free products (EUDR), to ensure that wood and paper products come from responsibly managed sources. When forest workers combine current capabilities in real-time monitoring, data collection and compliance verification in the field with Aarnibot’s ability to surface relevant information and instruction, they can be sure they are meeting environmental and social standards efficiently.
In conclusion
As generative AI steps out of the office and into the living world, forestry offers a powerful proving ground for what “AI in the wild” can achieve. Rather than replacing human expertise, the technology promises to amplify it, bridging data and environmental stewardship in real time. Tools like Aarnibot hint at a future where intelligent systems operate alongside the forest, turning dense, dynamic ecosystems into spaces of continuous learning and adaptive management. If responsibly developed and aligned with regulatory standards such as FSC, generative AI could become one of the most impactful allies in advancing both sustainable forestry and global climate resilience.
References / Notes
- FAO, Global Forest Resources Assessment 2020. Global Forest Resource Assessment 2020
- Meta, Using Artificial Intelligence to Map the Earth’s Forests, Meta Sustainability
- Corona, P. (2025). Integrating artificial intelligence to support systemic advances in silviculture. Annals of Silvicultural Research
- Holzinger, A., Schweier, J., Gollob, C. et al. From Industry 5.0 to Forestry 5.0: Bridging the gap with Human-Centered Artificial Intelligence. Curr. For. Rep. 10, 442–455 (2024). https://doi.org/10.1007/s40725-024-00231-7
