Trump administration backs OpenAI in copyright training dispute

The Trump administration has stepped into a high-stakes copyright battle by filing court papers defending OpenAI's practice of training its large language models on copyrighted content without permission. The development signals where the federal government stands on a contentious issue reshaping the legal landscape around artificial intelligence.

The Government's Position#

In a 20-page brief submitted to support OpenAI in its legal clash with The New York Times, the administration argued that restricting how AI systems learn from copyrighted works would damage American competitiveness. The government stressed that maintaining leadership in AI innovation requires freedom to develop and refine these technologies, referencing an executive order signed by President Trump earlier this year that prioritizes U.S. AI advancement.

The Central Legal Question#

Modern large language models—the engine behind ChatGPT, Claude, Gemini, and similar systems—rely on vast training datasets drawn from published books, news articles, and other written materials. Publishers contend this practice violates copyright law when done without licensing or compensation. The core dispute hinges on whether such training qualifies as "fair use," a legal doctrine that permits certain unlicensed uses of copyrighted work in specific circumstances.

The government's brief contends that overly strict interpretations of fair use principles could stifle innovation. According to the filing, the question of whether AI training is "transformative" enough to fall under fair use protection remains central to resolving these cases.

Momentum Favors Tech Companies#

Recent judicial decisions have largely sided with AI developers on the training question itself. When Anthropic was ordered to pay $1.5 billion to writers whose works trained its models, the penalty stemmed from the company's use of illegal book-piracy services, not from the act of training on copyrighted material. Judge William Alsup distinguished between the two, suggesting that LLM training parallels how humans learn from reading—an implicit endorsement of the practice's legitimacy.