AI and Copyright Law: Finding the Right Balance

The recent lawsuit filed by The New York Times against OpenAI and Microsoft over its artificial intelligence system has brought the issue of copyright infringement by AI into focus. As someone who has fielded questions on this topic before, I believe this is fundamentally more a matter of copyright law than technology. This lawsuit will probably draw clearer distinctions between the two domains. ?

Personally, I have never been a proponent of expansive copyright laws that seem to benefit large media companies at the expense of consumers and up-and-coming artists (writers, filmmakers, visual artists). The notion that any inspiration derived from published works constitutes infringement is particularly problematic to me. As a society, we are constantly exposed to and influenced by information around us, often in ways we don't even realize or remember. Should fleeting inspiration really be grounds for legal consequences years later?

“But the use of this technology also presents a possible existential crisis for the news industry, which has struggled to find ways to replace the revenue it once generated from its profitable print products. “

I do believe AI developers should pay licensing fees to access certain content, just as any other business would. However, news publishers face revenue challenges tied to numerous industry changes, not just technological advancements. Declining public trust in polarized news coverage deserves as much scrutiny as any technological disruption when analyzing the media's struggles — for an in-depth assessment of such problems, I recommend James Bennet’s piece in the Economist 1843 Magazine, "When the New York Times Lost Its Way." ?

Additionally, while publishers demand recompense when AI systems summarize their content, the same publishers do not pay any form of royalties to the victims featured in their true crime reporting, which certainly drives traffic to their sites. If we consider copyright principles comprehensively, such inconsistencies stand out. ?

Questions around derivative works also illustrate the complexities of this debate. If fiction authors build on themes and characters from myths and legends in the public domain, should their works also lose copyright protection? Inspiration has murky borders that our legal framework is ill-equipped to navigate. ?

There is no doubt AI systems raise copyright concerns that technology companies must address responsibly. However, as we balance those interests, we should be wary of arguments that may limit access to information mainly for those unable to pay. The early promise of the internet as an open repository of human knowledge relies on getting this balance right. Legal outcomes here will shape much more than profits.