Inventorship in AI-Assisted Inventions: Designing an Experiment to Shape Case Law

TEXT START: The latest improvements in artificial intelligence (AI) raise new challenges for intellectual property laws, particularly concerning the inventorship issue in AI-assisted inventions - that is, those in which AI is used in the inventive process.

B.1 THE DISSECTION

This paper proposes a deliberate regulatory capture operation disguised as neutral legal scholarship. The author acknowledges that current IP law cannot keep pace with AI's productive capacity, then proposes a solution that amounts to: "let's design the test cases to produce favorable precedents." The framing—"identifying the most effective methods for measuring human contribution"—reveals the actual agenda. This is not neutral legal science. It is a prescription for how to construct a fiction that preserves human inventorship rights in an era where the creative contribution increasingly belongs to the machine.

The author openly admits the problem: natural persons are legally required to be inventors, but AI is increasingly doing the inventive work. Their proposed "experiment" is not an experiment at all—it is litigation strategy. You don't "design experiments to shape case law." You file cases and let courts decide. What this paper describes is selecting which test cases to bring, controlling the facts to maximize favorable outcomes, and engineering judicial precedent. That is advocacy. The academic veneer is a delivery mechanism for the conclusion already reached.

B.2 THE CORE FALLACY

The paper assumes the human contribution problem is solvable through better measurement. This is the central error. Under the Discontinuity Thesis, the issue is not that we lack the methodological sophistication to measure "how much" a human contributed versus an AI. The issue is that the contribution is structurally migrating to AI and the measurement problem is permanently intractable at scale. No rubric—no matter how carefully "designed" through stakeholder consultations—will preserve human inventorship as AI systems become more capable. The paper treats this as an engineering problem requiring better instruments. It is a structural problem requiring acknowledgment of systemic displacement.

The inventorship requirement exists to connect economic value to human agency. When AI does the work, no measurement methodology restores that connection—it merely obscures the transfer.

B.3 HIDDEN ASSUMPTIONS

1. Human inventorship is normatively desirable and worth the institutional engineering required to preserve it. The paper never interrogates this premise. It assumes preserving human "inventor" status is a goal worth pursuing through deliberate case law manipulation.

2. Jurisdictions will cooperate in harmonizing inventorship standards through a coordinated "experiment." This assumes coordination is possible at scale. It is not. Different jurisdictions have different interests, and the nations that lead in AI development will face the strongest pressure to expand machine-eligible inventorship to capture IP value. The paper's proposed harmonization is a fantasy of institutional coordination that ignores competitive dynamics.

3. The "contribution" being measured is real and separable. You cannot measure what cannot be cleanly attributed. AI systems that suggest directions, complete circuits, identify novel compounds, or optimize parameters are not tools—they are co-creators. The paper assumes you can draw a bright line between "AI assistance" and "AI invention" when the reality is a continuum that will continue to shift toward AI autonomy.

4. The legal system can be engineered to stay ahead of AI development. The paper explicitly acknowledges this process "will not keep pace with the rapid development of AI." This is an admission that their solution is perpetually lagging, yet they propose it as a remedy. This is lag defense theater.

B.4 SOCIAL FUNCTION

This is transition management masquerading as legal scholarship. Its function is to:

• Provide intellectual cover for institutions (patent offices, courts, corporations) that need a framework to continue processing AI-assisted inventions without confronting the underlying displacement
• Signal to the legal community that "someone is working on this problem" to reduce anxiety
• Create a false sense that the human-centric IP system can be surgically adjusted rather than replaced
• Justify continued human-centric IP ownership structures that route value to existing rights-holders rather than AI developers

The paper is a document of institutional self-preservation. It exists because the legal profession, patent offices, and corporate IP departments need to believe the current framework is adaptable. The "experiment" proposal is institutional reassurance dressed in academic language.

B.5 THE VERDICT

The paper diagnoses a real problem—the legal system has no coherent answer for AI-driven invention—but proposes a solution that is both dishonest about its nature (advocacy, not neutral scholarship) and structurally inadequate to the displacement it acknowledges. Under the Discontinuity Thesis, the "human contribution" question has no stable answer that preserves human inventorship at scale. The patent system routes ownership and liability. When AI systems are the true inventors, no measurement methodology makes humans the inventors. The paper's "experiment" will produce case law that reflects the outcome that was designed in—human inventorship is preserved through selective filing and favorable fact patterns, not through principled legal reasoning.

The actual function: Buying time for existing IP holders while the structural displacement accelerates. This is hospice care dressed as preventative medicine.