A coalition of major American newspapers headed by the New York Times has escalated legal pressure against artificial intelligence company OpenAI, filing a motion in Manhattan federal court seeking sanctions for what they characterise as deliberate deception about the company's technical capabilities. The group, which includes the New York Daily News, alleges that OpenAI made false statements regarding its ability to search its language models for evidence that millions of news articles were improperly incorporated into the training data for ChatGPT, the AI chatbot that has become central to the broader debate over intellectual property rights in the emerging artificial intelligence sector.
The core accusation centres on OpenAI's handling of discovery obligations in litigation. According to the newspapers' court filing, OpenAI explicitly told the court that searching its large language models for copyrighted material would be technologically unfeasible and would violate user privacy protections. Simultaneously, the newspapers contend, the company was conducting exactly these searches, a contradiction they characterise as a deliberate attempt to obstruct justice. This alleged concealment extends to what the newspapers describe as the deletion or deliberate obscuring of billions of ChatGPT conversations that could have provided evidence of the company's misuse of news content.
What makes this development particularly significant is the scale of the claimed deception. The newspapers argue that OpenAI had already performed searches for their content before any of them formally initiated litigation, suggesting the company's later denials of technical capability were knowingly false rather than the result of genuine limitations. An employee of OpenAI subsequently testified under oath that the company had "performed multiple searches for News Plaintiffs' content," directly contradicting the corporation's earlier statements to the court. This contradiction between institutional positions and employee testimony has given the newspapers substantial grounds for their sanctions motion.
The legal consequences the newspapers are seeking extend beyond simple fines. They are requesting that the court impose attorney's fees upon OpenAI and make a judicial finding that the company's chat logs demonstrate it misused the newspapers' copyrighted works. Such sanctions represent a significant escalation in the litigation strategy and could substantially increase the financial and reputational costs to OpenAI of defending itself against these claims. The motion underscores growing frustration among news organisations with what they view as tech companies' cavalier disregard for intellectual property rights.
This particular dispute originated with a lawsuit filed by the New York Times in 2023, charging that OpenAI and Microsoft, which has invested billions in the AI startup, utilised millions of articles from the newspaper and other outlets without authorisation to train ChatGPT. The case represents one component of a much broader legal landscape in which content creators across multiple industries have launched coordinated challenges against AI developers. Authors, visual artists, photographers, and music labels have all initiated similar actions against companies including Anthropic, Meta Platforms, and OpenAI, each arguing that their creative work was incorporated into AI training datasets without consent or compensation.
The underlying issue speaks to fundamental questions about how artificial intelligence systems should be trained in the digital age and whether traditional copyright protections remain adequate safeguards for creative works. The newspapers argue that their journalism constitutes protected intellectual property that should not form part of AI training data without negotiated licensing agreements and appropriate compensation. OpenAI and other AI developers have contended that such training falls within fair use doctrine, a legal framework that permits certain uses of copyrighted material without permission under specific circumstances. This philosophical divide has not yet been resolved by the courts.
OpenAI has not publicly responded to the latest allegations, with a company spokesperson declining immediate comment on the sanctions motion when approached by media outlets. This silence is notable given the seriousness of the accusations and the high-profile nature of the litigation. The company's reluctance to immediately counter the newspapers' claims may reflect an assessment that public statements at this stage could complicate its legal position, or it may simply indicate that senior leadership is still formulating an appropriate response.
For Malaysian and Southeast Asian readers, this legal battle carries significant implications. The outcomes in American courtrooms will likely establish precedents that influence how AI companies worldwide approach the sourcing and use of copyrighted material. If the New York Times and allied newspapers succeed in demonstrating that OpenAI engaged in court deception and obtain substantial damages, this could establish a powerful disincentive for other AI developers to approach copyright protection casually. Conversely, if OpenAI successfully defends itself, it may embolden other technology companies to pursue similarly aggressive AI training methodologies.
The case also highlights the vulnerability of media organisations in the digital economy. News outlets across Southeast Asia, many of which already struggle with digital distribution and advertising revenue challenges, face the prospect that their content could be incorporated into AI systems that subsequently compete with their editorial operations. The outcome of this American litigation could determine whether regional publishers have legal recourse if their content is used without permission to train local or international AI systems.
Furthermore, the allegations of court deception, if proven, would raise serious questions about corporate accountability and the rule of law in the technology sector. The notion that a prominent company deliberately concealed its technical capabilities from a court, while simultaneously conducting the very activities it claimed were impossible, represents a troubling pattern that extends beyond copyright disputes into broader governance concerns. Such behaviour, if established, undermines the integrity of judicial processes and suggests that technology companies believe they can operate under different standards than traditional industries.
The upcoming developments in this case will likely hinge on whether the federal judge finds the newspapers' allegations of deception sufficiently credible to warrant sanctions, and whether OpenAI can effectively challenge these claims. If sanctions are imposed, the company faces not only financial penalties but also the serious risk that unfavourable findings regarding its conduct could prejudice its defence on the underlying copyright merits. The stakes continue to escalate for both sides as this crucial test case proceeds through the American legal system.
