തങ്ങളുടെ വാർത്തകൾ അനുമതിയില്ലാതെ എഐ പരിശീലനത്തിനായി ഉപയോഗിച്ചെന്ന് ആരോപിച്ച് ഓപ്പൺ എഐക്കെതിരെ യുഎസ്എ ടുഡേ കോടതിയിൽ. കോപ്പിറൈറ്റ് ലംഘനം ചൂണ്ടിക്കാട്ടി 250 മില്യൺ ഡോളറിലധികം നഷ്ടപരിഹാരമാണ് ആവശ്യപ്പെട്ടിരിക്കുന്നത്. മാധ്യമ സ്ഥാപനങ്ങളും സാങ്കേതിക കമ്പനികളും തമ്മിലുള്ള പകർപ്പവകാശ തർക്കങ്ങളിൽ നിർണ്ണായകമാകുന്ന കേസായി ഇത് മാറുകയാണ്.
USA Today Co has initiated a lawsuit against OpenAI over alleged unauthorized use of its copyrighted articles. The complaint claims OpenAI used hundreds of thousands of articles for training its AI models. USA Today Co seeks damages exceeding $250 million along with an injunction to stop the infringement. This case reflects a larger legal struggle between media publishers and technology companies regarding copyright use.
The growing use of artificial intelligence in news, research and everyday searches has created a difficult question for the media industry: can technology companies use journalism to build commercial AI systems without first obtaining permission from the publishers who produced it? A new lawsuit filed by USA Today Co against OpenAI has brought that dispute into sharper focus. The media corporation and its several newspapers have sued the AI firm, claiming that it has violated copyright laws by copying hundreds of thousands of articles and other materials to train the large language models for ChatGPT.
The lawsuit, which was filed in the US federal court in Manhattan, demands more than $250 million in damages and cessation of the copyright infringement. The allegations have not yet been proven in court, while OpenAI, in turn, did not give a reply to the complaint immediately after the lawsuit, as stated in the report.
This lawsuit is one of the numerous legal cases that are being solved in court regarding the use of copyrighted materials in training AI models. It has several implications that go beyond one particular company, touching upon the ways of obtaining information for AI systems, paying for original reporting and copyright laws in general.
Below are four major questions that underlie the controversy.
Can AI companies use news articles without publishers' permission?Question one revolves around whether OpenAI had the legal right to use the information outlined in the complaint filed by USA Today Co to train its AI model.
News organisations employ many efforts to come up with original news. The reporters conduct various activities such as gathering information, interviewing sources, and even examining documents to prove their findings. Publishers argue that this work represents intellectual property, even when articles can be accessed online.
According to the lawsuit, OpenAI copied hundreds of thousands of articles and other materials belonging to USA Today Co and used them in AI training without permission.
The court will need to look at what material OpenAI used, how it obtained it and how it was used. Just because a news article is available online does not mean it can be copied or used for any purpose. The key question is whether OpenAI violated the publisher's copyright or whether its use of the material is allowed under copyright law. The court's decision could influence how AI companies collect and use news articles in the future.
Does training ChatGPT on copyrighted material qualify as fair use?A second important issue is that of fair use. It is a doctrine in US copyright law which allows the use of copyrighted material without obtaining prior permission from the copyright owner.
The AI firms which are engaged in copyright litigation cases have argued that using the existing copyrighted materials to train a model is an instance of fair use since the machine learns from the patterns of the data and produces new output. The publishers and rights holders have disputed this argument and maintained that such widespread use of their work for AI development should not be deemed legal.
Fair use depends upon individual circumstances. The court takes into account the purpose of the use, the character of the copyrighted work, the extent of the use, and its effect on the market value of the copyrighted work.
Those questions can become complicated when AI training involves large collections of material. The court may need to consider whether the way the content was used is sufficiently transformative, as well as the commercial context and any effect on the publisher's ability to benefit from its reporting.
The USA Today lawsuit does not establish that OpenAI violated copyright law. Instead, it adds another set of claims to a broader legal debate over how existing copyright rules apply to generative AI.
Should publishers receive compensation when AI systems use their journalism?The dispute also raises a financial question: if news articles help companies develop commercial AI products, should publishers be paid for providing that material?
Conventional journalism depends on generating income through advertisements, subscriptions, licensing, etc. The gathering of news and stories may prove costly, especially where the stories require investigative work, specialised knowledge and fact-checking. The publishers fear that the AI product might get to use the content without having to pay for it.
Moreover, there are concerns that the answers generated by the AI technology might be sufficient to keep the user from having to visit the site.
The scale of that impact can vary. AI tools may introduce readers to publishers' work in some circumstances, while replacing visits to sources in others. The outcome depends on how a product operates, what information it provides and whether it directs users to the material it draws upon.
USA Today Company is demanding over $250 million as damages along with injunctions to prevent the claimed infringement from continuing. However, the complaint does not settle the issue of whether there is a legal entitlement to compensation for any payments due under the agreement.
The licensing agreements provide one option for reaching a commercial solution and allow the developers of AI technology to bargain for access to certain materials. What will happen in future is unknown.
Could the case change how AI companies build their models?Fourthly, can the lawsuit affect the way AI developers acquire training data and make deals with content owners?
The lawsuit against OpenAI is not an isolated case where technology companies have been accused of copyright violations. Authors and media companies have sued companies like Anthropic and Meta Platforms. The New York Times, among other well-known authors and news companies, has sued OpenAI and Microsoft in separate lawsuits in Manhattan's Federal court.
The suits vary in their allegations and evidence, but the outcome of one of them won't determine the rest. However, court rulings might give us some insights into acceptable use of copyrighted material.
If the courts find against such training techniques, then the developers would be forced to get a licence, alter the method by which they collect data or restrict access to certain materials. Where certain uses are held to be within the bounds of fair use, there would be greater clarity on what is acceptable as training material.
As far as the publishers are concerned, the impact would have an effect on both negotiating power as well as the valuation of their archives. For the technology companies, the decisions would have a bearing on the cost of development and material that can be used for training the next generation of models.
The main problems still stand unsolved - the permission that AI developers need, fair use of training and the necessity for payment to publishers.