A group of prominent newspapers, anchored by the New York Times and New York Daily News, has escalated their copyright dispute with OpenAI by seeking court sanctions for what they characterise as sustained dishonesty during legal proceedings. The filing in Manhattan federal court on Thursday alleges that OpenAI made false statements regarding its technical capabilities, specifically denying it possessed methods to search its large language models for copyrighted content while simultaneously concealing that it had already performed such searches before litigation began.
The core dispute centres on whether OpenAI improperly utilised millions of newspaper articles to train ChatGPT, its widely-used artificial intelligence chatbot. According to the newspapers' court submission, OpenAI claimed that searching its systems for their copyrighted material would be infeasible, burdensome, and potentially invasive of user privacy. Yet internal company communications and employee testimony suggest OpenAI possessed the technical means to conduct these searches and had indeed exercised that capability.
The newspapers further contend that OpenAI engaged in document destruction or obscuration by deleting or rendering unsearchable billions of ChatGPT conversations potentially containing evidence of copyright misuse. This allegation adds a layer of obstruction to the underlying infringement claims, as it suggests deliberate elimination of materials that might have proven central to establishing how the company trained its systems. The plaintiffs have consequently requested that the court impose sanctions, award attorney fees, and make a judicial finding that OpenAI's chat logs demonstrate misappropriation of their copyrighted works.
OpenAI's response to these accusations remained pending at the time of reporting, with company representatives declining immediate comment on the motion. The silence stands in contrast to earlier statements in which the company disputed technical feasibility of conducting the searches now alleged to have been performed. This strategic posture in the ongoing litigation reflects the complexity and stakes involved in resolving one of the tech sector's most significant copyright disputes.
The original lawsuit, initiated by the New York Times in 2023, expanded the scope of copyright actions challenging artificial intelligence development practices. Alongside the Times and Daily News, the case involves Microsoft, the world's largest investor in OpenAI, as a co-defendant. The fundamental accusation remains that both entities unlawfully incorporated copyrighted journalistic material into training datasets without authorisation or compensation, thereby enabling the creation of a commercial product that generates value while creators received nothing.
For Malaysian and regional readers, this dispute carries particular relevance as Southeast Asian media organisations increasingly grapple with the same questions regarding AI companies' use of their content. The outcome in this high-profile American case may establish precedents that ripple through international copyright law and influence how artificial intelligence companies operate globally, including in markets such as Malaysia, Singapore, and Indonesia.
The legal confrontation between established media institutions and AI developers forms part of a broader reckoning across creative industries. Authors, visual artists, music labels, and other copyright holders have similarly sued OpenAI, Anthropic, Meta Platforms, and other technology firms for allegedly misusing their material in training systems. These lawsuits collectively represent an attempt to establish that generative AI companies cannot simply absorb vast quantities of creative work without permission or payment, a principle that could substantially reshape AI development economics.
Ian Crosby, the New York Times' lead counsel, articulated the newspapers' perspective forcefully in a statement accompanying the motion. He characterised OpenAI's conduct as representing over two years of sustained misrepresentation directed at the newspaper plaintiffs, the Daily News plaintiffs, the general public, and the judicial system itself. The claim extends beyond technical disputes about capability to encompass broader questions about corporate honesty and good faith in litigation.
The apparent contradiction between OpenAI's courtroom statements and internal evidence emerged through employee testimony, which revealed that the company had "performed multiple searches for News Plaintiffs' content," directly contradicting earlier denials. Such discrepancies typically draw judicial scrutiny and can result in adverse inferences, where courts assume facts asserted by the lying party are established against them. This dynamic potentially strengthens the newspapers' underlying copyright claims beyond the sanctions motion itself.
The timing and manner of OpenAI's alleged deception matter considerably for determining appropriate remedies. If the company deliberately misrepresented its capabilities rather than genuinely discovering capabilities during discovery, courts view this more severely as deliberate obstruction. Similarly, if conversations were deleted after litigation commenced, the intentionality becomes clearer. These factual questions will likely consume significant judicial attention before sanctions are imposed.
The broader implications for artificial intelligence governance remain profound. If courts determine that AI companies systematically misrepresent their capabilities and destroy relevant evidence, this may prompt regulatory responses across jurisdictions. Policymakers in Malaysia and throughout Southeast Asia watching this dispute may feel emboldened to impose stricter requirements on AI companies operating in their territories regarding transparency and content sourcing practices.
The case also reflects an asymmetry in power between established technology companies and traditional media organisations. While OpenAI possesses substantially greater resources and technical sophistication, the newspaper plaintiffs command considerable reputational platforms and legal expertise. This balance creates an unusually competitive litigation environment compared to cases where individual creators challenge corporations.
