On Nov. 28, 2025, the United States Patent and Trademark Office (USPTO) issued revised guidance on inventorship for AI-assisted inventions, rescinding its 2024 guidance. The update does not impose new restrictions on AI-driven research. Instead, it restores traditional inventorship principles centered on human conception and confirms that AI systems are treated as tools, not inventors. 

For chemical and materials companies that rely heavily on AI for compound screening, formulation optimization, and process modeling, the guidance reduces uncertainty and allows existing inventorship practices to continue with modest refinements rather than structural change.

What the USPTO Clarified

• There is no special inventorship standard for AI-assisted inventions

• Only natural persons may be named as inventors

• Inventorship turns on human conception, not tool usage

• AI systems are treated as research tools similar to modeling software or laboratory equipment

• Joint inventorship analysis applies only when multiple humans are involved

What Changed From the 2024 Guidance

The USPTO withdrew its prior reliance on joint inventorship principles for AI-assisted inventions where only one human contributor was involved. Because AI systems are not persons and cannot be joint inventors, the USPTO confirmed that the earlier framework was unnecessary. The revised guidance returns inventorship analysis to established Federal Circuit law.

Practical Implications for Chemical and Materials Companies

AI-driven discovery remains safe
Using AI to screen compounds, predict material properties, optimize formulations, or model processes does not affect inventorship.

Human judgment remains decisive
Inventorship belongs to those who define the problem, interpret AI outputs, and select the specific solutions reflected in the claims.

Claim-focused inventorship reviews remain essential
Inventorship should continue to be evaluated claim by claim, particularly in multidisciplinary teams.

Global filing strategy still matters
U.S. applications must name human inventors, and priority claims may be jeopardized if foreign filings list only AI systems.

What Companies Should (and Should Not) Do Now

Do

• Continue using AI aggressively in R&D

• Document human decision-making tied to claim conception

• Review inventor alignment across jurisdictions

Do Not

• Name AI systems as inventors

• Assume AI output alone establishes conception

• Overcorrect by narrowing inventorship unnecessarily

Bottom Line

The USPTO’s revised guidance simplifies inventorship analysis for AI-assisted inventions and reduces risk for chemical and materials portfolios. AI accelerates invention, but inventorship remains grounded in human conception. The practical effect is continuity, not disruption.