OpenAI Warns US to Let It Train on Copyrighted Material or China Wins AI Race

OpenAI has urged the US government to give the company unrestricted access to copyrighted material to train its AI models and points to China as the reason why it should escape copyright laws.

OpenAI is asking the US government to make it easier for AI companies to learn from copyrighted material, citing a need to “strengthen America’s lead” globally in advancing the technology.

Open AI, the start-up behind ChatGPT and DALL-E, submitted the suggestions to the U.S. government on Thursday as part of President Donald Trump’s upcoming “AI Action Plan.”

President Trump ordered his administration advisors to formulate such a plan earlier this year and has asked for input from the private sector, government, and academia in the U.S.

In a 15-page letter to the US government on Thursday, OpenAI urged the federal government to enact a series of “freedom-focused” policy ideas that allow the companies to train its models on copyrighted material — including an approach that would no longer compel American AI developers to “comply with overly burdensome” state-level AI bills in the U.S.

OpenAI says that if it is not able to train its models on copyrighted material, China will take the lead in the AI race.

OpenAI described DeepSeek’s latest model, R1, as a “noteworthy” advancement, highlighting China’s expanding AI ambitions and the increasing competition between the two countries.

“While America maintains a lead on AI today, DeepSeek shows that our lead is not wide and is narrowing,” Chris Lehane, OpenAI’s vice president of global affairs, writes in a letter to the Office of Science and Technology Policy.

OpenAI did not specify how it would achieve this but stated that promoting “fair use” policies and reducing intellectual property restrictions could help “[protect] the rights and interests of content creators while also protecting America’s AI leadership and national security.”

The use of copyrighted material in AI training remains highly controversial, as many companies continue to train models on human-created content without consent or compensation.

AI companies have faced numerous lawsuits from photo agencies, news outlets, and artists which allege unlawful use of their copyrighted content.

Image credits: Header photo licensed via Depositphotos.