Patent Issues Relating To Typhoon-Prediction Algorithms Built By PhilIPpine Climate Tech Startups.
1. Core Legal Problem: Can Typhoon-Prediction Algorithms Be Patented?
Under the Philippine Intellectual Property Code (Republic Act No. 8293):
- A patent must be:
- Novel
- Inventive (non-obvious)
- Industrially applicable
🚫 Major Barrier: Software Exclusion
- Computer programs “as such” are NOT patentable
- Pure algorithms (e.g., storm prediction models, AI forecasting logic) are treated as:
- mathematical methods
- abstract ideas
👉 Therefore, a typhoon prediction algorithm alone cannot be patented.
✅ Exception (Critical for Startups)
If the algorithm:
- Produces a technical effect, or
- Is embedded in a technical system (e.g., satellite data processing, sensor networks, disaster-warning infrastructure),
→ It MAY be patentable as a computer-implemented invention
2. Key Patent Issues for Climate-Tech Startups
(A) Algorithm vs Technical Application
- Startup builds:
- AI model predicting typhoon paths → ❌ not patentable
- AI integrated with real-time weather sensor hardware → ✅ possibly patentable
(B) Data & Model Training Issues
- Ownership of:
- datasets (satellite, oceanic data)
- trained models
- These often fall under copyright / trade secrets, not patents
(C) AI Inventorship Problem
- Philippine guidelines (2025):
- AI cannot be an inventor
- Human developer must be identified
(D) Enforcement Weakness
- Very few patent litigation cases in the Philippines
→ weak precedent and uncertainty for startups
3. Important Case Laws (Detailed Analysis)
Even though none directly involve climate algorithms, these cases shape how such technologies would be treated.
1. Boehringer Ingelheim v. Suhitas (2021)
Facts:
- Patent for hypertension drug (telmisartan)
- Defendant imported generic version before patent expiry
Legal Issues:
- Patent validity
- Scope of infringement
- Exceptions under public health law
Held:
- Patent was valid, novel, and enforceable
- Unauthorized importation = infringement
Relevance to Typhoon Algorithms:
- Confirms:
- Strong protection if a patent is valid
- Startups must ensure clear novelty + claims drafting
👉 Lesson: If a climate-tech patent is granted (e.g., hardware-integrated system), courts will strictly enforce exclusivity
2. Merck Sharp & Dohme Corp. v. Zuellig Pharma Corp.
Facts:
- Dispute over pharmaceutical patent enforcement
Held:
- Supreme Court upheld patent validity and exclusivity rights
Legal Principle:
- Patent rights are strict property rights
Relevance:
- Climate-tech startups can:
- License their forecasting systems
- Prevent competitors from copying patented technical implementations
👉 Reinforces commercialization value of patents
3. Phillips Seafood Philippines Corp. v. Tuna Processors, Inc. (2023)
Facts:
- Patent on food processing technique
- Dispute on whether infringement occurred
Key Doctrine:
- Doctrine of Equivalents
- Patent protection limited to claims wording
Held:
- Courts cannot extend protection beyond claims
Relevance to Algorithms:
- If a startup patents:
- “Typhoon prediction system using X sensors + Y model”
Competitors can:
- Slightly modify architecture → avoid infringement
👉 Lesson: Precise claim drafting is critical for algorithm-based systems
4. E.I. DuPont de Nemours (Philippine jurisprudence reference)
Principle:
- Patent law balances:
- Disclosure to public
- Exclusive rights to inventor
Relevance:
- Climate startups face dilemma:
- Patent → disclose algorithm
- Trade secret → no protection if leaked
👉 Strategic choice:
- Keep AI model as trade secret
- Patent only technical infrastructure
5. Creative Minds v. SynthoArt (AI-related case, 2024)
Facts:
- AI-generated content used commercially without authorization
Held:
- Protection lies in:
- training data
- model design
Relevance:
- Typhoon AI models:
- Protection may lie in dataset + architecture, not output
👉 Important for:
- Climate startups using satellite/weather datasets
6. General Doctrine from IPOPHL AI Guidelines (2025)
(Not a case but quasi-legal authority)
Key Rules:
- AI inventions must:
- show technical contribution
- clearly disclose human inventorship
- avoid claiming abstract models alone
Relevance:
- Typhoon prediction system must:
- demonstrate real-world technical impact
- not just predictive accuracy
4. Application to Typhoon-Prediction Startups
Scenario Analysis
✅ Patentable Example:
A startup develops:
- AI model + ocean buoys + satellite integration
- Real-time early warning system
✔ Patentable because:
- Technical system
- Industrial application (disaster mitigation)
❌ Not Patentable:
- Pure ML model predicting cyclone intensity using historical data
✘ Falls under:
- algorithm
- mathematical method
5. Key Legal Risks
(1) Idea Theft
- Algorithms are hard to protect
- Reverse engineering risk
(2) Weak Litigation Ecosystem
- Few cases → uncertainty
(3) Claim Drafting Challenges
- Too broad → rejected
- Too narrow → easy to bypass
(4) Overlap with Public Data
- Weather data often:
- public domain
- non-proprietary
6. Strategic Recommendations for Startups
Hybrid IP Strategy:
- Patent
- hardware systems
- sensor integration
- data-processing pipelines
- Trade Secret
- ML models
- training datasets
- Copyright
- software code
7. Conclusion
Patent law in the Philippines creates a significant barrier for typhoon-prediction algorithms because:
- Pure algorithms are not patentable
- Protection depends on technical implementation
- Case law emphasizes:
- strict claim interpretation
- strong enforcement once granted
- importance of novelty and technical contribution
👉 Therefore, climate-tech startups must move beyond “just algorithms” and frame their innovations as technical systems solving real-world problems.
RELATED Blog
comments