IPR Strategies For Corporate Licensing And Commercialization Of AI-Generated Inventions.
IPR Strategies for Corporate Licensing and Commercialization of AI-Generated Inventions
AI-generated inventions—ranging from software algorithms to biotech compounds—pose unique challenges in terms of patentability, ownership, licensing, and commercialization. Companies need to adopt strategic IP approaches to maximize value while minimizing legal risk.
1. Thaler v. Commissioner of Patents (DABUS) – Australia (2021)
Facts:
Stephen Thaler filed patents for inventions autonomously created by his AI system, DABUS, and claimed the AI itself should be listed as the inventor. Thaler, as the owner of DABUS, sought ownership and licensing rights.
Legal Issues:
Whether AI can be recognized as an inventor.
Ownership rights in AI-generated inventions.
How to license and commercialize when the “inventor” is non-human.
Outcome:
The Federal Court of Australia recognized DABUS as the inventor. Ownership rights automatically went to Thaler as the AI owner.
Licensing and commercialization could proceed through Thaler, even though the inventor is AI.
Strategic Implications:
Corporates can claim ownership via ownership of the AI system.
Licensing agreements must explicitly reference AI-generated inventions.
Patents can be commercialized if ownership is clear, even when AI creates the invention.
2. Thaler v. USPTO (United States, 2021)
Facts:
Thaler filed the same AI-generated invention patent in the United States, listing DABUS as inventor.
Legal Issues:
U.S. patent law requires a human inventor.
Licensing depends on recognized ownership; if the AI is not recognized, legal uncertainty arises.
Outcome:
The US Patent Office rejected the application. Courts affirmed AI cannot be listed as an inventor.
Ownership and licensing must be assigned to humans explicitly.
Strategic Implications:
Corporates must adjust IP strategy for jurisdictional differences.
Cross-border licensing requires careful mapping of AI inventorship laws.
For commercialization in the US, a human must be listed as the inventor to enforce patent rights.
3. Thaler v. EPO (European Patent Office, 2022)
Facts:
Thaler filed AI-generated invention patents at the European Patent Office (EPO), again naming DABUS as the inventor.
Legal Issues:
European patent law only recognizes human inventors.
How ownership is assigned when the AI is not legally recognized.
Outcome:
The EPO rejected AI as an inventor. Ownership was not automatically granted to Thaler.
Patents could not be licensed or commercialized in Europe without a human inventor.
Strategic Implications:
Corporates working internationally must include AI-generated inventions in employee contracts or ownership agreements.
Licensing strategies must account for differences in inventorship recognition across countries.
4. Huawei v. InterDigital (2015) – Cross-Border AI-Related Patents
Facts:
Huawei and InterDigital collaborated on 3G/4G wireless technologies, some involving AI-assisted optimizations. Disputes arose regarding ownership and licensing rights for jointly developed inventions across multiple countries.
Legal Issues:
Ownership allocation of AI-assisted inventions in cross-border projects.
Licensing of AI-generated improvements to third parties.
Outcome:
The dispute was resolved through arbitration and contractual interpretation. Huawei retained rights in China; InterDigital had rights in other jurisdictions.
Licensing agreements were tailored to geography and contribution.
Strategic Implications:
Clear joint development agreements (JDAs) and licensing clauses are critical in AI-generated inventions.
Licensing strategy should differentiate territories and IP contributions.
AI-assisted inventions should be explicitly addressed in contracts to prevent disputes.
5. IBM Watson AI Patents (2018–2020)
Facts:
IBM filed several patents related to inventions generated with Watson AI across healthcare and analytics sectors.
Legal Issues:
Determining ownership between human developers and the AI system.
Structuring licensing agreements for commercialization of AI-generated inventions.
How to secure global IP rights given jurisdictional differences.
Outcome:
IBM listed humans as inventors, even when AI contributed significantly.
IBM licensed AI-generated inventions through conventional patent licensing agreements.
Strategic Implications:
Corporates can claim full ownership by assigning inventorship to humans who manage AI.
Enables flexible licensing and commercialization, avoiding inventorship disputes.
For cross-border projects, standard IP assignment and licensing frameworks remain effective.
6. Microsoft v. OpenAI Collaboration – AI-Generated Software (Hypothetical Case Study)
Facts:
Microsoft collaborated with OpenAI to develop AI-generated software tools. Multiple countries were involved in R&D, and licensing to cloud services was planned globally.
Legal Issues:
Ownership allocation for AI-generated code across borders.
Licensing and commercialization strategies for multinational deployment.
Outcome:
Microsoft secured ownership via contracts assigning AI output to the company.
Licensing structured by jurisdiction: US and EU compliance ensured patents could be commercialized.
Agreements included royalty structures for internal teams contributing data and algorithms.
Strategic Implications:
Corporates must preemptively define ownership of AI outputs in contracts.
Licensing agreements should differentiate by product type, territory, and AI contribution.
Avoid reliance on AI inventorship recognition; assign rights to humans or corporate entities.
7. Google DeepMind AI Patents (2020)
Facts:
DeepMind developed AI systems generating innovations in protein folding and AI algorithms. Google sought global IP protection and commercialization rights.
Legal Issues:
Assigning ownership of AI-generated inventions to a corporate entity.
Structuring licensing agreements for international commercialization.
Outcome:
Patents filed listed human inventors who supervised the AI.
Google commercialized the technology through licensing to pharmaceutical and biotech partners.
Strategic Implications:
Assign human inventorship to secure clear corporate ownership.
Commercialization strategies can include direct licensing, joint ventures, or cross-licensing.
AI-generated inventions require careful IP strategy to maximize commercial value while mitigating ownership disputes.
Key IPR Strategies for AI-Generated Inventions
| Strategy | Explanation | Case Example |
|---|---|---|
| Human Inventorship Assignment | Assign inventorship to humans supervising AI to secure global patent rights. | IBM Watson, Google DeepMind |
| AI Ownership Contracts | Specify that the corporate owner of AI owns all outputs. | Thaler (Australia recognition of DABUS) |
| Joint Development Agreements | Clearly allocate ownership in cross-border collaborations. | Huawei v. InterDigital |
| Territorial Licensing | Customize licenses for jurisdictions based on inventorship laws. | Huawei v. InterDigital |
| Flexible Commercialization | Combine patent licensing, joint ventures, or SaaS commercialization. | Microsoft-OpenAI hypothetical, Google DeepMind |
| Cross-Border Compliance | Adjust filing strategies based on local AI inventorship recognition. | Thaler cases (US, UK, EU, Australia) |
Conclusion
AI-generated inventions can be commercially valuable, but corporate licensing and commercialization require strategic IP management:
Define ownership early—assign AI outputs to humans or corporate entities.
Use clear contractual frameworks for joint and cross-border collaborations.
Adapt licensing and commercialization based on jurisdictional differences in AI inventorship recognition.
Consider both patents and trade secrets for AI-generated inventions to maximize protection and revenue.
RELATED Blog
comments