TimesOfAI’s Post

Here's why the New York Times lawsuit against OpenAI is an interesting case: The NYT filed a copyright infringement lawsuit against OpenAI in December 2023. The case is still ongoing. The major allegations against OpenAI are: 1. Training: OpenAI has scraped thousands of NYT articles to build its model. 2. Output/memorization: The output of OpenAI is a possible derivative of NYT articles, and it even provides excerpts from NYT articles when someone asks pointed questions like, "Tell me what NYT has said about ...." While OpenAI is pushing back by claiming 'fair use' (as with every copyright lawsuit), it will be interesting to see what happens if this goes to trial. If the NYT wins this case, the consequences could drastically change the AI landscape. - OpenAI may not allow queries that reproduce NYT content. - OpenAI may be forced to delete NYT articles from the training data (which may not be practical but represents a worst-case scenario). - OpenAI might need to pay millions of dollars in damages (the amount per license multiplied by the number of articles). - Other publishers may follow the NYT's lead and file lawsuits against OpenAI, further complicating the situation. Let's wait and see what happens. #openai #fairusage

Well, this is interesting. Authors Guild (yes, that one that made fair use law in the Google Books case...but not in the way they wanted) is suing OpenAI alleging copyright infringement for training its AI model on plaintiffs' works. You may recall, Authors Guild previously sued HathiTrust for scanning books (and lost on fair use) and Google for scanning books (and lost on fair use). This time, they're suing OpenAI for...scanning books.

Interesting that the plaintiffs insist the ability to create an accurate summary of their works is proof the works are in the training data -- but generative AIs are not sources of truth (see: the lawyer who got fabricated case law). GenAIs are just sources of pattern recognition and probability stats. The referenced works may very well be part of the training data; I don't know. But the ability to summarize the plot is not proof of that: it's proof that numerous other people have described the same works in the same way, to the point that the model recognizes the pattern.

BREAKING After a class action suit on behalf of nonfiction authors, the NY Times is the first major American media organization to sue OpenAI and Microsoft for copyright infringement: The lawsuit filed in Manhattan « contends that millions of articles published by The Times were used to train automated chatbots that now compete with the news outlet as a source of reliable information » « Defendants seek to free-ride on The Times’s massive investment in its journalism » « using The Times’s content without payment to create products that substitute for The Times and steal audiences away from it » More than ever, IP is crystallizing major issues at the core of technological change. What balance should be struck between protecting the legitimate interests of publishers vital to our democracy and enabling innovation and competition? Some plead for competition which is no surprise. We've already seen this with the advent of the Internet and the Safe Harbor clause for hosting providers. But the stakes appear even higher now, as it's our collective memory and accumulated intelligence that is being sucked up here! https://lnkd.in/ei2pT-X2

You cannot assign copyright to Visual Studio code. So why should you be able to assigne copyright to Copilot? Copyright needs to be assigned to a natural person (i.e., one human) or a legal person (i.e., a bunch of humans). #copyright #copilot #ai

The United States Copyright Office is undertaking a study of the copyright law and policy issues raised by artificial intelligence (“AI”) systems. To inform the Office's study and help assess whether legislative or regulatory steps in this area are warranted, the Office seeks comment on these issues, including those involved in the use of copyrighted works to train AI models, the appropriate levels of transparency and disclosure with respect to the use of copyrighted works, and the legal status of AI-generated outputs https://lnkd.in/eJp_jW9g

In an interesting move, the Copyright Office is taking public comments regarding #AI and #copyright. The inquiry lists over 30 questions like: - How is the use of this technology currently affecting or likely to affect creators, copyright owners, technology developers, researchers, and the public? - Under what circumstances would the unauthorized use of copyrighted works to train AI models constitute fair use? - Should Congress establish a new federal right, similar to state law rights of publicity, that would apply to AI-generated material? I'm looking forward to reading what creators, developers, and attorneys have to say about this. https://lnkd.in/gypi_xew

🔔 Breaking News Alert! 🔔 Did you hear about the latest controversy in the tech world? Google, the global tech giant, has found itself in the midst of a legal battle, accused by publishers for allegedly promoting pirated textbooks. 📚💻 The publishers claim that Google's search algorithms are favoring websites that offer illegal copies of their textbooks, thereby infringing on copyright laws. This issue raises several pertinent questions about the role of search engines in promoting or discouraging illegal content. What's your take on this? Do you believe search engines should be held accountable for the content they promote, or is it up to the user to discern legal from illegal? Let's start a conversation! Share your thoughts below. 👇 #Google #LegalBattle #TechNews #CopyrightInfringement #LinkedInDiscussion