University of Tennessee Sues Anthropic Over Neural Network Technology

A lawsuit has been filed by the University of Tennessee against a rival AI startup, claiming the company illegally used its patented technologies for neural network development.

The University of Tennessee is suing artificial intelligence startup Anthropic, alleging that the company infringed on its intellectual property associated with its technology for creating neural networks. According to the suit filed by the University of Tennessee against Anthropic, the AI company utilized some patented technologies without authorization when designing its own AI models, which have benefited greatly from its faculty research into cutting-edge technology used in today’s AI models.

The dispute adds to a growing trend of intellectual property wars occurring within the technology industry as AI startups race to create ever-more advanced language models and machine learning systems. Though numerous recently-filed AI lawsuits have concerned copyrighted training data, the new University of Tennessee vs. Anthropic dispute instead involves the underlying technologies powering neural networks, the basis for many AI systems.

Modern AI technologies are powered by neural networks. Based on the layout of the human brain, neural networks allow machines to find patterns, process information in the form of language, create images and make complex arguments or logical calculations. Their use ranges widely from personal AI assistants, to translation tools, to search engines and medical diagnostic devices, and they power the search results on Google and the language capabilities of smart devices or chatbots, to name just a few.

The university is asserting that patented research originating with the university’s faculty and researchers has been utilized by Anthropic without its having been licensed properly, seeking an award of damages as well as licensing terms and/or an order prohibiting future use.

Anthropic is a significant player in the AI space, producing what is arguably one of the leading AI companies in the space. Founded in 2019, the San Francisco based company has attracted $2 billion in investment from major companies like Amazon and Google, and it now operates with a $4 billion valuation as the company ramps up production and sales for its latest generation AI tools that compete directly with OpenAI's GPT series.

Many of the methods that make up artificial intelligence originated in universities. Academic researchers and institutions have developed a substantial amount of technology over the decades that have been sold and then commercially exploited, leading to intellectual property challenges similar to this one as they become much more lucrative.

The lawsuit is one in a string of similar lawsuits being filed as technology companies face increased scrutiny over AI-related patent and intellectual property cases. Legal uncertainty surrounds technology innovation. However, patents and copyrights can also apply to AI training data, though the legal precedent for them in the AI realm is not well-defined in many instances and courts are still attempting to rule on how many aspects of the technology apply.

According to legal experts, patent cases are distinct from AI training data copyright cases in that patents deal with inventions or technical processes, rather than the copying of specific material, and they believe a successful patent case could inspire similar claims and encourage future legal protections for AI innovations developed at research universities.

The case is expected to draw close attention from academia and the AI industry because if the University of Tennessee prevails, other universities may be spurred to take aggressive measures to patent its AI-generated inventions and seek payment when their inventions become the backbone for commercialized AI technology.

The future relationship between research institutions and the AI industry may hinge in large part on the case in which the University of Tennessee is fighting Anthropic for intellectual property rights in an area of science that is developing so rapidly, creating both an increased appetite for research and potential disputes as that research becomes integrated into commercial AI products.