AI Information & Transparency

PlantDr AI was created to make botanical science, plant care guidance, and disease diagnosis accessible to everyday gardeners and plant enthusiasts. We believe that deploying computer vision and machine learning responsibly requires absolute transparency about what our models can and cannot do.

This document details how our botanical models work, the inputs they analyze, why uncertainty exists in plant pathology, and how user privacy is safeguarded.

Artificial intelligence powers several key features within PlantDr AI:

1. Species Identification: Trained on extensive botanical taxonomy datasets, our computer vision models analyze leaf venation, leaf margins, floral morphology, stem structure, and growth habit to suggest species classifications with confidence percentages.

1. Botanical Health & Symptom Analysis: Our diagnostic pipeline examines visual symptoms of distress—including leaf chlorosis (yellowing), necrosis (browning/dieback), powdery fungal growth, rust pustules, viral mosaic patterns, and insect stippling.

1. Health & Vitality Scoring: The system computes an overall vitality score and categorizes probable underlying causes across fungal, bacterial, viral, environmental, and pest vectors.

1. "Ask AI Botanist & Pathologist" Assistant: A specialized conversational assistant that answers plant care questions, analyzes custom symptoms, and explains biological recovery options.

Inputs Processed:

• High-resolution photographs of leaves, flowers, stems, and soil surfaces.
• User-provided symptom observations (e.g., watering frequency, indoor vs. outdoor location, recent fertilizer applications).
• Camera ambient light telemetry for lux estimation.

Architecture & External Processing Infrastructure:

• Secure Application Architecture: Client devices communicate exclusively with hardened Axxa Media application services. Service credentials and security tokens are strictly guarded server-side and never exposed in client applications.
• Provider Neutrality & System Evolution: Automated botanical diagnostics, computer vision classifications, and generative reasoning are delivered via third-party AI technology services and external processing infrastructure. Axxa Media does not operate exclusively with any single artificial intelligence provider; underlying models, architectures, and external infrastructure providers may change, be upgraded, or transition over time to advance diagnostic precision and system resilience.
• Zero Foundation Model Training on User Photos: Under commercial enterprise terms with our third-party artificial intelligence service providers, user-submitted plant photos and conversational queries are processed statelessly and are NOT retained or used to train or refine public foundation machine learning models.

Botanical pathology is inherently complex. Users must understand why AI diagnoses can never be guaranteed:

• Symptom Mimicry: Different pathogens produce nearly identical symptoms. For instance, overwatering root rot and severe underwatering both present as drooping, yellowing leaves. Nitrogen deficiency can look identical to early root rot chlorosis.
• Early-Stage Invisibility: Fungal mycelium or bacterial vascular blockages may cause plant decline days before outward symptoms become visible on leaves.
• Image Quality Factors: Motion blur, shadow, overexposure, or photographing the wrong part of the plant (e.g., photographing a leaf when the disease is in the roots) can degrade diagnosis reliability.
• Multiple Concurrent Pathogens: A plant may simultaneously suffer from spider mites, root rot, and fertilizer burn.

PlantDr AI provides decision-support tools, not definitive guarantees. You remain solely responsible for any physical interventions you perform:

1. Always Test Chemical Treatments: When applying fungicides, insecticides, or home remedies, test a single leaf before spraying the whole plant.
2. Verify Environmental Factors: Check physical soil moisture with your finger or a soil probe before assuming a plant is diseased.
3. Observe Toxicity Precautions: Always check plant toxicity if children or pets share your home.
4. Consult Experts for High-Value Specimens: For mature trees, valuable landscaping, or commercial crops, consult a certified local arborist or agricultural extension service.

We continuously refine our botanical pipelines. If you encounter a clear species misidentification or an inaccurate disease diagnosis, please report it via the in-app "Refine Diagnosis" tool or email [email protected] with the photo and plant record ID. For account support or credential security issues, contact [email protected].