Skip to main content

OpenAI highlights briefs supporting fair use for AI training

30 OCTOBER 2026·3 MIN READ·5 SOURCES

OpenAI says experts and leaders shared perspectives in court briefs supporting fair use, with arguments that AI training serves a different purpose from the works it learns and does not provide substitute access to them. The voices it highlights include authors, journalists, academics, accessibility advocates, legal experts and business leaders.

OpenAI highlights briefs supporting fair use for AI training

Key takeaways · 4

  • 01

    OpenAI says court briefs supporting fair use include perspectives from authors, journalists, creators, academics, accessibility advocates and business leaders.

  • 02

    The AI Progress brief argues that AI training has a different purpose from copyrighted works, does not provide the public substitute access to them and does not cause cognizable market harm.

  • 03

    The Foundation for American Innovation argues that capabilities enabling new analysis, software and productive work have national-security value.

  • 04

    Wall and Siskind said AI helps them create more or extend their work into new modes, while Siskind uses it in her business and Wall says it lowers costs.

A broad set of fair-use supporters

OpenAI says experts and leaders shared perspectives in court briefs supporting fair use[1]. The group it describes includes authors, journalists, creators, academics, accessibility advocates, legal experts, librarians, economists and business leaders[1]. OpenAI says restrictions on fair-use training would affect institutions building AI and people using it[1].

How the briefs distinguish training

The AI Progress brief argues that AI training serves a different purpose from the copyrighted works it uses, does not give the public substitute access to those works and does not cause cognizable market harm[1]. A separate argument describes foundation-model training as an effort to build a general-purpose tool others can adapt, rather than to reproduce the works used in training[1]. Professor Matthew Sag argues that training copies and model outputs are not meaningfully similar under realistic conditions[1]. OpenAI characterizes The New York Times’ lawsuit as asking the court to disregard legal history and fair use in pursuit of what OpenAI calls an undeserved payday[1].

Claims about work and public impact

The briefs OpenAI cites also connect fair-use training to potential practical and economic effects. The Foundation for American Innovation argues that capabilities enabling new analysis, software and productive work have national-security value[1]. The U.S. government argues that excluding large language model training from fair use would hamper scientific and creative progress and hinder American prosperity and economic mobility[1]. Restrictions on fair-use training would affect institutions building AI and people using it, according to OpenAI[1]. Author Kevin Kelly describes each AI model as a new medium that gives authors capabilities to create texts and ideas and to reason[1].

Examples from creative businesses

OpenAI’s account includes examples from Wall and Siskind[1]. It says Wall and Siskind argued that AI helps them create more and extend their work into new modes of expression[1]. Siskind uses AI across her business, including administration, developing ideas and finding new clients, according to the brief[1]. Wall says AI helps him manage his music business, lower costs and spend more time composing[1]. Public Knowledge separately argued that fair use has protected image search and full-text book search[1]. Journalist Ron Suskind and law professor Brian L. Frye argued that OpenAI’s use of newspaper articles did not affect the plaintiffs’ ability to sell consumers access to those articles[1].

For professionals, the briefs frame fair use as a question with implications for both the ability to build AI systems and the ways people may use them at work. The arguments are positions advanced by OpenAI and the cited contributors; they do not establish how a court will decide the issue.

Why it matters
Daily session

Bring this to your team — one session a day, built for your industry.

kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.

For teams

How this developed

  1. 31 October 2026

    OpenAI says experts and leaders have shared perspectives in court briefs supporting fair use, including arguments about AI training’s purpose, effects and potential benefits.

  2. 31 October 2026

    This update clarifies that the Wikimedia Foundation asked for permission to file an amicus brief; the evidence does not say the brief had already been filed, and it adds details about the request and proposed brief’s length.

  3. 31 October 2026

    The updated account specifies that Wikimedia says providers of retrieval-augmented generation should follow each platform’s or publisher’s licensing requirements, and warns that a ruling applying fair use carelessly could threaten free knowledge, open-source technology and access to information.

  4. 30 October 2026

    Wikimedia urges limits in AI copyright case

Sources

AI fluency, one session a day, built for your work.