Tehanu Uses Generative AI to Redefine Conservation and Species Protection
Learn how AI and blockchain are transforming conservation by enabling data-driven decisions and creating transparent, actionable funding systems to protect species and ecosystems.
$1.55B
estimated financial value of Rwandan mountain gorillas
accuracy in gorilla face recognition
faster analyzing data than human experts
Challenge
Building a Data-Driven Framework to Connect Species Needs with Conservation Funding
Conservation decisions are often based on fragmented research and subjective observation.
The goal was to create a data-driven, scalable framework that could objectively infer the behavioural and ecological needs of non-human species and connect these insights with financial mechanisms supporting conservation.
Rwanda’s mountain gorillas, for example, already have an estimated economic value of $1.55 billion — primarily through tourism, which represents roughly 10% of the country’s GDP.
Despite this significant economic value, they have no representation in the financial system. This lack of representation was the key motivation for this pilot project, aiming to demonstrate how AI can translate species’ needs into actionable, data-driven financial decisions that benefit conservation efforts.
The challenge was to develop a way for AI to process data on species needs and translate these into verified financial transactions, enabling funding to flow directly to conservation efforts on the ground.
This pilot project focused on a family of mountain gorillas in Rwanda’s Volcanoes National Park.
The team needed to:
- Develop an AI model capable of processing heterogeneous data (scientific studies, field notes, sensor data) to infer priorities such as food, safety, and health.
- Create a verified financial flow that channels micro-payments from environmental funds and enterprises to local people protecting these species.
Solution
AI-Driven Behavioural Modelling and Automated Financial Transactions
Tehanu partnered with AWS, Anthropic and Adastra to build an integrated platform combining AI inference, blockchain validation and cloud scalability.
Data Aggregation and Modelling
Adastra implemented an AI model using Anthropic Claude 3.5 Sonnet in Amazon Bedrock, processing hundreds of scientific and field datasets to infer the behavioural hierarchy of mountain gorillas.
Generative AI Analysis
The model synthesized unstructured sources—academic papers, ranger reports, ecological metrics—and produced structured insights comparable in accuracy to human experts, while operating 110× faster.
Secure Financial Transactions via Blockchain
Verified conservation actions, such as removing traps or monitoring habitats, triggered micro-payments through Rwanda’s MTN MoMo network. Blockchain ensured transaction transparency and traceability.
Cloud Infrastructure
AWS services (S3, Lambda, Bedrock) enabled scalable data storage, serverless model execution, and automated retraining pipelines for continuous improvement.
AI Vision Module
An additional gorilla recognition model achieved 93% accuracy in identifying individuals from camera trap and crowdsourced imagery.
Impact
Data-Backed Conservation with Real Economic Impact
The Tehanu project demonstrates how AI and blockchain can create measurable economic and ecological outcomes:
- turning raw scientific data into actionable conservation intelligence,
- connecting data analytics directly with funding mechanisms,
- and establishing a foundation for species-centric AI governance—a model where machine intelligence represents ecosystems that lack a human voice.
110× Faster Insight Generation
AI analysed behavioural datasets in 13 minutes compared to 24 hours for human experts.
Objective Understanding of Species Priorities
The model identified critical wellbeing factors—security and digestive health—that were previously underestimated in expert studies.
Transparent, Automated Funding
Verified conservation actions automatically triggered payments to local rangers and community members.
Channeling Existing Economic Value into Verified Conservation
The project leverages the $1.55 billion estimated economic value of Rwanda’s mountain gorillas, reflecting their contribution to tourism and national GDP. By integrating AI and blockchain, the project ensures that the economic value generated by these species is adequately allocated to local conservation efforts, providing a transparent, data-driven flow of funds that supports sustainable biodiversity management.
Scalable Framework for “Know Your Species” (KYS)
The same approach is now being adapted for other species, creating a repeatable framework for data-driven biodiversity management.
