ABSTRACT

This article examines how conventional intellectual property regimes are challenged by AI- generated discoveries. By analyzing global trends in patents, copyrights, and trademarks, it explores emerging issues like criminal liability and ethical reforms. Ultimately, it provides tactical suggestions for legislators navigating the c

omplex and evolving interplay between the development of AI and legal rights.

KEYWORDS

Artificial intelligenc

e, intellectual property rights, copyright, patent

INTRODUCTION

“When machines can write and paint, we stop asking what AI can do and start asking what we value.” Since machine learning now enables autonomous creative and imaginative outputs, the integration of artificial intelligence into industries like healthcare and finance has upended conventional intellectual property models. Predictive modeling is one way AI improves efficiency, but its ability to conduct independent discovery calls into question the fundamental legal theory that ties intellectual property rights to human action.

This article looks at jurisprudential conflicts that arise within the domains of copyright, patents, and trademarks. The “inventorship” requirements in patent law and the “human authorship” requirement in copyright, both of which are presently being tested by AI- generated works, are crucial to this approach. Furthermore, the study addresses the liability and unique character hurdles that AI poses to trademark enforcement, finally assessing the suitability of current legal frameworks in the face of non-human innovation.

WHAT ARE INTELLECTUAL PROPERTY RIGHTS?

Intellectual Property Rights (IPR) are the legal safeguards guaranteed to individuals or organizations for “creations of the mind.” IP regulates the proprietary interests in intangible creations of art and inventions that possess unique economic or societal value. The categories of Intellectual property rights are as follows:

• Patent: It protects inventions.
• Copyright: It safeguards creative works such as artistic and literary.
• Trademark: It safeguards names and logos used by brands.
• Trade Secret: Safeguards proprietary data (recipes, formulas).

WHAT IS ARTIFICIAL INTELLIGENCE?

Artificial Intelligence (AI) is the branch of computer science devoted to creating machines that can carry out operations that have historically required human intelligence. This encompasses the ability to learn on one’s own, think logically, and recognize intricate patterns.

KEY COMPONENTS

• Machine Learning: Systems that use dataset analysis to maximize performance.
• Deep Learning: A sophisticated type of neural networks with numerous layers (hence the term “deep”) that enables the system to process extremely complicated input, such as natural language or facial recognition.
• Natural Language Processing (NLP): NLP enables machines to comprehend and produce human language. Used in chatbots, language translation, sentiment analysis, and more.

INTELLECTUAL PROPERTY RIGHTS IN ARTIFICIAL INTELLIGENCE

In 2026, the relationship between Artificial Intelligence (AI) and Intellectual Property Rights (IPR) is not merely a theoretical discussion; rather, it is a rapidly evolving legal reality that is characterized by new required rules and “human-centric” requirements.

This is a more thorough analysis of the current situation:

1. Is AI capable of being an “Author” or “Inventor”?

The prohibition that AI cannot be an author or creator has been firmly established by the majority of international authorities, including the US, EU, and UK.

• COPYRIGHT: AI-generated works are typically denied registration by the US Copyright Office and various other international organizations. A work must have “substantial human creative control” in order to be protected.
• PATENTS: Following decisions such as Thaler v. Vidal, courts in the United States, the United Kingdom, and Europe have ruled that an AI cannot be designated as an inventor. The only “natural person” is eligible. New USPTO rules for 2025-2026, however, make it clear that humans can use AI as a tool as long as they “significantly contributed” to the invention.

2. The transition to Mandatory Licensing

“Free training” is coming to an end. By early 2026, the industry was moving toward a licensing model because of high-profile lawsuits like NYT v. OpenAI have pushed the industry toward a licensing model.

• Opt-In/Opt-Out: The EU’s new 2026 standards and those of the music industry (such as Sony, Universal) center on “Opt-In” systems, in which AI developers are required to pay for the usage of copyrighted data in their training sets.
• Remuneration Rights: Some areas are implementing “statutory remuneration,” which is comparable to radio royalties and pays creators automatically when an AI model incorporates their work.

3. Trademark and “Digital Parasitism”

In 2026, AI-generated branding will encounter two significant legal obstacles:

• Infringement Liability: The user, not an AI, is usually responsible for “unintentional infringement” if the AI designs a logo that unintentionally imitates a protected brand.
• Digital Identity: There is a growing trend in courts safeguarding “digital likeness.” Nowadays, it is considered a major violation of the Right to Publicity or trademark- like personality rights to use AI to imitate someone else’s voice or mannerisms without that person’s consent.

4. Key regulatory benchmarks for 2026
   • Liability: who bears responsibility?
     1. The User: Held accountable if they purposefully ask the AI to imitate a protected brand (for example, by requesting a “Disney- style logo”).
     2. The Developer: Responsible if the AI is a “black box” that repeats data protected by copyright without being asked to do so, or if they neglected to put protections in place after being informed of infringement.
   • Transparency: Is it Real?
     1. Watermarking: In order to demonstrate that AI-generated media is not human- made, new rules require it to have visible labels (for deepfakes) and invisible metadata (C2PA/provenance standards).
     2. The EU AI Act (August 2, 2026) mandates that “High-Risk” AI (used in recruiting, legislation, or infrastructure) keep stringent activity records and demonstrate that their training data was managed ethically.
     3. Provenance: To trace a file’s origin from the time of creation to the point of dissemination, Digital “certificates” are being appended to it.

        

(This blog was written by Ojashvi Verma, a fifth-year law student at City Academy Law College. This blog examines the evolving global legal framework, highlighting issues of liability, licensing, and regulatory reforms such as the EU AI Act in response to AI-driven innovation.)