AgriCount Privacy Policy
AgriCount (“AgriCount,” “we,” “our”) offers digital tools for agricultural monitoring, yield estimation, and risk analysis using artificial intelligence and computer vision. We serve individual farmers, cooperatives, producer organizations, agribusinesses, financial institutions, and sustainability programs in Latin America and globally. Privacy, data sovereignty, and the protection of information at the farm level are fundamental principles of AgriCount’s design and governance. This Privacy Policy explains how we collect, use, protect, and manage personal and operational data when you use our mobile applications, desktop software, websites, and related services (the “Services”).
1. Fundamental Principles
AgriCount is governed by the following principles:
-
Data Ownership: Farmers and businesses retain ownership of their information.
-
Purpose Limitation: The data is used solely for agronomic analysis, productivity estimation, risk management and service improvement.
-
Anonymization by Design: No farmer, farm, or person can be identified in the model training processes or in aggregate analyses.
-
No Data Marketing: We do not sell or market personal or identifiable data.
-
Aggregate Use: Model training and comparative analyses are performed using only anonymized and aggregated data.
2. Information We Process
2.1 Data Provided by Users
• Photographs of crops, trees, plots and production units
• Property boundaries and geospatial polygons
• Agronomic observations and production metrics
• Metadata for projects, batches, and operations
• Contact information (when user accounts exist)
2.2 Automatically Generated Data
• Pod counts, yield estimates, and confidence levels
• Productivity and risk indicators
• Model performance statistics
• Technical and security records
3. Use of Images and Artificial Intelligence
AgriCount uses images to:
Perform local or on-device analysis (detection, counting, and classification).
Improve the accuracy of models by training with irreversibly anonymized and aggregated datasets.
Guarantees:
No personal identifiers are stored in the training data.
The images are not labeled with names, household details, or precise locations.
Aggregation thresholds are applied that prevent re-identification.
Models learn statistical patterns, not individual information.
The original images are never sold, licensed, or used for advertising or profiling purposes.
4. Legal Basis for Processing
5. Data Sharing
9. Use by Companies, Cooperatives and Institutions
The data can only be shared:
-
With the express authorization of the user (certification, financing, research).
-
In aggregated and anonymized form for:
-
Scientific research
-
Sustainability reports
-
ESG Analysis
-
-
With infrastructure providers under strict confidentiality agreements.
AgriCount does not share identifiable data with advertisers, data brokers, or social networks.
For medium and large operations:
-
AgriCount acts as the Data Processor.
-
The client acts as the Data Controller.
-
Data Processing Agreements (DPAs) regulate:
Confidentiality
-
Purpose
-
International transfers
-
Governance of machine learning model training
-
10. Minors
6. Information Security
AgriCount implements:
-
Encryption in transit and at rest
-
Access controls and audit logs
-
Secure cloud and on-premises architectures
-
Logical separation of training environments
-
Periodic security assessments
7. International Transfers
AgriCount is not designed for use by anyone under the age of 16 and does not intentionally collect data from children or teenagers.
11. Governance of Model Training
-
Only anonymized and aggregated data is used.
-
No inferences are preserved at the individual level.
-
No facial or biometric recognition processes are performed.
-
Training systems are separate from operating systems.
When data is processed outside the country of origin, the following apply:
-
Safeguards equivalent to the GDPR
-
Contractual clauses compliant with the LGPD
-
Regional accommodation where possible
-
Principles of minimization and pseudonymization
12. Modifications
-
This Policy may be updated to reflect regulatory or technological changes. The current version will be available at www.agricount.com
8. Rights of Data Subjects (Latin America)
AgriCount respects the rights established in:
-
Colombia – Law 1581 (Habeas Data)
-
Brazil – LGPD
-
Peru – Law 29733
-
Ecuador – LOPDP
-
Dominican Republic – Law 172-13
-
Data protection laws of Central America
Users can request:
-
Access to your data
-
Rectification
-
Elimination
-
Opposition or limitation of treatment
-
Withdrawal of consent for model training
-
Data portability
Contact:
13. Contact
-
Email: info@agricount.com
Depending on the jurisdiction, AgriCount processes data based on:
-
User consent
-
Contract execution
-
Legitimate interest (model improvement, security, aggregate analysis)
-
Compliance with legal obligations