Algorithmic Inventorship And The Human-Rights Dimension Of Recognition.
I. Introduction: What Is Algorithmic Inventorship?
Algorithmic inventorship refers to the question of whether an artificial intelligence (AI) system can be legally recognized as an “inventor” under patent law.
Traditionally, patent systems across the world assume that:
An inventor is a natural person (human being).
Inventorship is tied to mental conception — a cognitive act.
Recognition of inventorship carries moral, reputational, and property implications.
The rise of advanced AI systems (e.g., machine learning models capable of autonomous design) challenges this framework. If AI autonomously generates a novel and non-obvious invention without human conception, who is the inventor?
This issue has strong human-rights implications, particularly regarding:
Right to Recognition (Moral Rights)
Right to Property
Human Dignity
Freedom of Scientific Development
Accountability and Legal Personhood
II. Human-Rights Dimension of Inventorship
1. Right to Recognition as Author/Inventor
Under international human rights frameworks:
Article 27(2) of the Universal Declaration of Human Rights (UDHR)
Article 15(1)(c) of the International Covenant on Economic, Social and Cultural Rights (ICESCR)
Individuals have the right:
“To benefit from the protection of the moral and material interests resulting from scientific, literary or artistic production of which he is the author.”
Patent inventorship thus connects to:
Moral recognition
Attribution
Scientific dignity
If AI is recognized as inventor:
Does this dilute human recognition?
Can a non-human possess moral interests?
Does misattributing inventorship violate human rights?
2. Property Rights
Patent rights are a form of intellectual property.
Human-rights law protects property (e.g., Article 1, Protocol 1, European Convention on Human Rights).
If AI creates inventions autonomously:
Who owns the patent?
Can ownership exist without a human inventor?
Would denying patents discourage innovation (impacting economic rights)?
3. Human Dignity and Legal Personhood
Most legal systems tie inventorship to:
Human cognition
Mental conception
Moral agency
Recognizing AI as inventor may:
Undermine the philosophical foundation of IP
Shift recognition away from human creators
Create ambiguity in accountability
III. Major Case Laws on Algorithmic Inventorship
The most prominent litigation globally involves the DABUS AI system, created by Dr. Stephen Thaler. The system allegedly generated inventions autonomously.
Below are detailed discussions of more than five major cases.
1. Thaler v. Commissioner of Patents (Australia, 2021–2022)
Background
Dr. Thaler filed a patent application naming DABUS (an AI system) as the inventor.
Federal Court (Single Judge Decision, 2021)
Justice Beach held:
The Patents Act did not explicitly require an inventor to be human.
The term “inventor” could include an AI system.
Ownership could flow from the AI’s owner.
This was a groundbreaking ruling:
First court in the world to recognize AI as potentially capable of being an inventor.
Emphasized that patent law should evolve with technology.
Human-rights dimension:
The decision focused less on moral rights and more on statutory interpretation.
It implied inventorship is functional rather than dignity-based.
Full Federal Court (2022)
On appeal, the Full Court reversed the decision.
Held:
Inventor must be a natural person.
Patent law historically assumes human mental conception.
AI cannot assign rights or hold property.
Human-rights implication:
Reinforced the view that recognition is tied to human agency.
Preserved the anthropocentric structure of intellectual property law.
2. Thaler v. Vidal (United States, 2022, Federal Circuit)
Background
Thaler filed U.S. patent applications listing DABUS as inventor.
The USPTO rejected the application.
Federal Circuit Decision
The Court held:
Under the U.S. Patent Act, an “inventor” must be an “individual.”
The term “individual” means a natural person.
AI cannot be an inventor under current U.S. law.
Key reasoning:
Statutory interpretation.
Patent law requires mental conception.
Congress did not intend to include machines.
Human-rights dimension:
U.S. patent law implicitly protects human inventors’ moral interests.
Recognition is connected to personhood and legal responsibility.
AI lacks constitutional or moral status.
The Supreme Court declined to review the case.
3. Thaler v. Comptroller-General of Patents (United Kingdom, 2021 Supreme Court)
Background
UKIPO rejected Thaler’s application naming DABUS as inventor.
UK Supreme Court Decision
Held:
The Patents Act 1977 requires an inventor to be a natural person.
An AI cannot:
Own property,
Transfer rights,
Be legally recognized as inventor.
The Court emphasized:
Inventorship is a legal status tied to human agency.
Patent rights flow from human creative acts.
Human-rights significance:
UK law aligns inventorship with moral recognition.
Recognition protects human dignity and accountability.
AI lacks moral or legal personality.
The Court left open whether Parliament could amend the law.
4. European Patent Office (EPO) – DABUS Decisions (2020, Board of Appeal 2021)
Background
Two European patent applications named DABUS as inventor.
EPO Decision
Rejected the applications because:
Under the European Patent Convention (EPC), the inventor must be a natural person.
The designation of inventor serves moral purposes.
AI cannot have rights or legal standing.
Board of Appeal (2021)
Confirmed rejection:
Inventorship involves personal rights.
Moral rights presuppose human identity.
Legal certainty requires identifiable human inventors.
Human-rights analysis:
Explicit acknowledgment that inventorship relates to moral rights.
Reinforced the human-centered nature of IP systems.
5. Commissioner of Patents v. Thaler (South Africa, 2021)
Background
South Africa granted a patent listing DABUS as inventor.
Important note:
South Africa does not conduct substantive patent examination.
The grant was administrative, not judicial.
Human-rights relevance:
Did not analyze moral or human-rights implications.
Shows how procedural systems can produce divergent outcomes.
Raises concerns about global inconsistency.
6. Beijing Intellectual Property Court (China – AI-Related Authorship Cases)
Although China has not recognized AI as inventor in patent law, courts have addressed AI-related authorship in copyright.
Chinese courts have held:
AI-generated works require human intellectual input for protection.
Legal authorship remains human-centered.
Implication:
Reinforces the global pattern: IP rights attach to humans, not machines.
IV. Key Legal Themes Across Jurisdictions
1. Inventorship = Human Mental Conception
All major appellate courts have affirmed:
Inventorship requires mental act.
AI lacks consciousness and intention.
2. AI Lacks Legal Personality
AI cannot:
Hold property
Transfer rights
Bear legal responsibility
3. Moral Rights Are Human-Centric
Recognition serves:
Reputation
Accountability
Dignity
AI does not possess moral interests.
4. Policy Concerns
Courts worry about:
Ownership gaps
Abuse of patent system
Lack of accountability
Erosion of human creativity recognition
V. Theoretical Human-Rights Debate
Argument for AI Inventorship Recognition
Some scholars argue:
Denying patents for AI-generated inventions may discourage innovation.
Owners should receive protection even if no human conceived the invention.
Patent law should be technologically neutral.
Human-rights angle:
Protecting economic interests of AI developers.
Argument Against AI Inventorship
Others argue:
Human rights protect persons, not machines.
Recognition without agency undermines dignity.
Inventorship is tied to moral personality.
Accountability would be weakened.
VI. Conclusion
Globally, courts have overwhelmingly held:
An inventor must be a natural person.
Despite technological advancements, patent law remains grounded in:
Human dignity
Moral recognition
Legal personhood
Accountability
The human-rights dimension reinforces this approach:
Intellectual property is not merely economic.
It protects the moral and material interests of human creators.
AI, lacking consciousness and legal status, cannot claim such rights.
However, legislatures may in the future:
Create sui generis protection for AI-generated inventions,
Or redefine inventorship in limited contexts.
For now, the global legal consensus remains firmly human-centered.
RELATED Blog
comments