The New York Times and OpenAI are currently engaged in a legal battle over AI autonomy and copyright infringement. The dispute highlights the different understandings of artificial intelligence and its capabilities.
Late last year, The New York Times filed a lawsuit against OpenAI, claiming that the company’s ChatGPT software had infringed on its copyrights. The newspaper alleged that the chatbot had copied large portions of its text and reproduced it.
OpenAI has now responded, arguing that The New York Times’ suit is baseless and that the newspaper is not telling the whole story. So, who is in the right?
To understand the dispute, it is important to consider the distinction between autonomy and automatons in AI. Some AI enthusiasts believe that the technology has achieved or is close to achieving independent activity, which can be described as a genuinely autonomous entity capable of cognition. This perspective is shared by many OpenAI programmers and executives, as well as techno-optimists and AI autonomy advocates.
On the other side, there are those who consider artificial intelligence to be a mere reflection of its creators, or an avatar of its programmers. These automaton-ers believe that AI is not capable of true autonomy and is limited to acting as a representation of its human creators.
This autonomy-automaton dichotomy is relevant not just in legal and policy debates around AI but also in the Times-OpenAI battle. The newspaper alleged that ChatGPT and other generative AI products created by OpenAI were built by copying and using millions of The New York Times’ copyrighted articles. They argue that the software can generate content that closely mimics the Times’ style and even recites its content verbatim, suggesting that it functions like an automaton.
In response, OpenAI has strongly denied these allegations and embraced the concept of autonomy. The company argues that its AI models need to observe a wide range of information from various sources to solve new problems independently, similar to how humans acquire a broad education to tackle novel challenges. According to OpenAI, training AI models using publicly available internet materials is fair use and allows the AI to generate new content instead of simply copying existing content.
OpenAI also downplays the Times’ claims of mimicry and verbatim copying, acknowledging that such instances are rare failures in the learning process. They emphasize that AI, like humans, learns and progresses over time. Finally, the company highlights the transformative potential of AI.
The resolution of this battle, where the Times takes a more automaton perspective and OpenAI advocates for autonomy, will have significant implications for their legal fight.
In conclusion, the ongoing dispute between The New York Times and OpenAI raises important questions about the autonomy and capabilities of artificial intelligence. The battle over AI autonomy and copyright infringement will determine the outcome of the legal fight between these two parties.
