• 

Patent Issues For AI-Designed Modular Typhoon-Resilient Shelters.

1. Key Patent Issues

(A) AI as an Inventor

• Patent law traditionally requires a human inventor.
• If an AI system autonomously designs a shelter (e.g., optimal aerodynamic structure for typhoon winds), the question arises:
  👉 Who is the inventor? The developer? The user? Or the AI?

(B) Patent Eligibility (Subject Matter)

AI-designed shelters may involve:

• Mathematical models (wind simulations)
• Algorithms (optimization models)
• Structural outputs (physical shelter)

Issue:

• Are AI-generated designs abstract ideas or patentable technical inventions?

(C) Obviousness / Inventive Step

• AI systems can generate thousands of design permutations
• Courts may ask:
  👉 Would this design be obvious if AI tools are commonly available?

(D) Ownership & Authorship

• If multiple actors are involved:
  • AI developer
  • Engineer using AI
  • Construction company
    👉 Who owns the patent?

(E) Enablement & Disclosure

• Patent law requires clear disclosure
• AI models (especially black-box systems) make it difficult to explain:
  👉 How exactly was the design generated?

2. Important Case Laws (Detailed)

1. Thaler v. Comptroller-General of Patents

Facts:

• Stephen Thaler filed patent applications naming an AI system (DABUS) as the inventor.

Issue:

• Can an AI system be legally recognized as an inventor?

Judgment:

• The UK Supreme Court held:
  ❌ AI cannot be an inventor
  ✔ Only a natural person qualifies

Relevance:

• If an AI designs a typhoon-resistant shelter:
  • The patent must name a human inventor
  • Even if AI generated the design autonomously

Key Principle:

Inventorship is tied to legal personality, which AI lacks.

2. Thaler v. Vidal

Facts:

• Same DABUS system was used to file US patents.

Issue:

• Whether US patent law allows non-human inventors.

Judgment:

• The court ruled:
  ❌ Only humans can be inventors under US law

Importance:

• Reinforces global trend:
  👉 AI-generated shelter designs require human attribution

3. Alice Corp. v. CLS Bank International

Facts:

• Concerned patentability of computer-implemented inventions.

Issue:

• Whether abstract ideas implemented via software are patentable.

Judgment:

• Introduced the Alice Test:
  1. Is the claim directed to an abstract idea?
  2. Does it add an “inventive concept”?

Relevance to AI shelters:

• AI algorithms optimizing wind resistance may be seen as:
  ❌ Abstract mathematical models
• But:
  ✔ If tied to physical shelter structure, patentable

Key Insight:

Pure AI design logic ≠ patentable
AI + physical engineering application = potentially patentable

4. Diamond v. Diehr

Facts:

• Concerned software used in rubber curing.

Judgment:

• Allowed patent because:
  ✔ Software was applied to a physical process

Application:

• AI-designed shelters:
  ✔ Patentable if AI output is tied to real-world structural improvements

Principle:

Industrial application makes algorithms patent-eligible

5. KSR International Co. v. Teleflex Inc.

Facts:

• Addressed obviousness standard

Judgment:

• A patent is invalid if:
  👉 It combines known elements in a predictable way

Relevance:

• AI-generated shelter designs:
  • If AI merely combines known structural techniques:
    ❌ May be considered obvious
  • If it produces unexpected wind-resistance behavior:
    ✔ Likely patentable

6. EPO Decision J 8/20 (DABUS case)

Issue:

• Whether AI can be named as inventor under European law

Judgment:

• ❌ AI cannot be an inventor
• ✔ Inventor must be human

Additional Insight:

• EPO emphasized:
  👉 Legal accountability requires a human inventor

7. Naruto v. Slater

Facts:

• A monkey took a selfie; question was whether it could hold copyright.

Judgment:

• ❌ Non-humans cannot hold intellectual property rights

Relevance:

• Though a copyright case, it reinforces:
  👉 AI cannot own or be credited with IP

8. Bilski v. Kappos

Issue:

• Patentability of abstract processes

Judgment:

• Abstract ideas are not patentable

Application:

• AI design methods alone:
  ❌ Not patentable
• Structural shelter innovations:
  ✔ Patentable

3. Application to Typhoon-Resilient Modular Shelters

Patentable Aspects:

✔ Structural design (e.g., aerodynamic roofing)
✔ Modular assembly systems
✔ Material innovations
✔ AI-assisted engineering methods tied to physical output

Non-Patentable / Risky Areas:

❌ Pure AI algorithms
❌ Data models without physical application
❌ Designs obvious from existing engineering practices

4. Practical Legal Strategy

For patenting such shelters:

1. Human Inventor Positioning

• Clearly identify:
  👉 Engineer who guided or interpreted AI output

2. Claim Drafting

• Focus on:
  ✔ Structural features
  ✔ Performance improvements (wind resistance)
  ✔ Real-world deployment

3. Avoid Abstract Claims

Instead of:

• “AI system for optimizing shelter design”

Use:

• “A modular shelter structure comprising… designed to withstand wind speeds of X km/h”

4. Explain AI Role

• Include:
  ✔ Training data
  ✔ Design constraints
  ✔ Engineering validation

5. Conclusion

AI-designed typhoon-resilient shelters are patentable, but only if:

• A human inventor is identified
• The invention is technically grounded (not abstract)
• It shows non-obvious structural innovation

The major takeaway from global case law:

Patent law is evolving—but still firmly human-centered, even in AI-driven innovation.

RELATED Blog

Software Patenting

OpenAI Sued by Multiple Copyright Lawsuits in Indi...

Delhi High Court Rules in Favor of Svamaan in Trad...

US Supreme Court to Decide Whether AI-Generated Ar...

Intellectual Property Laws at Afghanistan

Intellectual Property Laws at Albania

Intellectual Property Laws at American Samoa (US)

Intellectual Property Laws at Angola

Intellectual Property Laws at Antigua and Barbuda

Intellectual Property Laws at Armenia