AI chatbots that replicate literary, artistic, and other creative works are having copyright issues, according to Bloomberg.
TakeAway Points:
- AI chatbots, such as ChatGPT, are in their pursuit to emerge as the most revolutionary technical force since the founding of the internet.
- The copyrighted writers, publishers, news organisations, and other parties whose works were utilised to train the massive language models of the chatbots are demanding a share of the earnings.
- The Bloomberg study is released at a time when The New York Times has already sued Microsoft and OpenAI for copyright infringement.
Generative AI’s Copyright Controversy
AI chatbots that mimic the works of poets, artists, and other creatives are running into copyright problems. According to a Bloomberg report, generative artificial intelligence systems such as ChatGPT, which aspire to be the most disruptive technological force since the internet’s inception, have ingested a vast amount of material, including millions of songs, beat poetry, draft contracts, movie scripts, photo essays, and novels from the 19th century.
Bloomberg explains that ChatGPT and other AI chatbots produce texts, graphics, and audio that can compete with those of a gifted human. But this is done by consuming previously produced information, looking for patterns in it, and then using those patterns to create new content.
In its quest to become the most disruptive technological force since the internet’s creation, these generative artificial intelligence systems have devoured millions of songs, beat poetry, draft contracts, movie scripts, photo essays, and novels from the 19th century, among other materials.
It seems that this thorough investigation of human history comes at a price. News outlets, authors, music publishers, and other companies whose copyrighted works were used to train the chatbots’ massive language models are requesting a share of the profits.
Conflicts in the Law and Fair Use
The legal landscape is currently navigating through at least 20 cases, primarily in California and New York federal courts, involving claims against AI companies for copyright infringement. High-profile cases include the New York Times suing OpenAI and Microsoft for up to $450 billion in damages, alleging the unauthorized use of its articles to train ChatGPT.
The Bloomberg story is released at the same time as The New York Times filed a lawsuit alleging copyright infringement against Microsoft and OpenAI. In this case, the NYT claims that the businesses that created ChatGPT and other well-known AI systems have unfairly used its written works.
The complaint, which was filed with the Federal District Court in Manhattan, claims that automated chatbots that were trained on millions of The Times articles are now directly competing with the news organisation as trustworthy sources of information.
Industry Reactions and Possible Fixes
While litigation continues, some AI companies are seeking to negotiate commercial terms with content creators. For instance, the Associated Press and Germany’s Axel Springer have reached licensing agreements with OpenAI, allowing their content to be used for training AI in exchange for compensation. These deals suggest a path forward that could balance the interests of copyright holders with the development of generative AI technologies. However, individual authors and smaller news organizations face challenges in securing similar agreements, highlighting the complexities of copyright negotiations in the AI era.