Technology news and discussion on TalkFuse.
Americans don’t have to choose between our AI future and ripping off copyright owners
The article argues that the US Department of Justice has taken an overly broad position in a major copyright dispute involving OpenAI and Microsoft. Several news organisations, including The New York Times, allege the companies used millions of articles to train AI systems without permission; the companies say this is fair use. The author says the DOJ’s support for broad commercial use without licensing or compensation risks weakening creators’ rights.
The DOJ says licensing costs could disadvantage smaller developers and hinder US AI leadership and national security. The article contrasts that argument with White House guidance and G20 principles that recognise copyright’s role and describe AI copyright issues as complex. It also says some AI start-ups favour licensing for more reliable access and less legal uncertainty, and notes that US copyright-related industries contribute more than $1 trillion to the economy.
In short
- DOJ backs broad fair use for commercial AI training.
- News organisations say millions of articles were used without permission.
- The dispute weighs AI development against creators’ rights and compensation.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Broad access to copyrighted works for AI training could help developers build useful systems, strengthen US competitiveness and support national security. Requiring licences for vast training datasets may impose costs and legal burdens that smaller developers struggle to meet, potentially concentrating innovation among large firms.
The case against
Using millions of articles without permission or compensation risks weakening creators’ ability to control and earn from their work, while licensing could give AI developers more reliable access and reduce legal uncertainty. Copyright-related industries make a substantial economic contribution, and recognising creators’ rights need not rule out AI development if workable licensing arrangements can support both.
Explained simply
Generative AI systems learn patterns from large collections of material, including written work, and use them to produce new responses. US copyright law gives creators rights over how their work is used, but whether training AI on published material needs permission is being disputed in court.
The New York Times and other news organisations have sued OpenAI and Microsoft, alleging that their articles were used without permission; the companies say this use is allowed as fair use. The US Department of Justice has weighed in, arguing that licensing requirements could raise costs for smaller developers and affect US competitiveness and national security.
The dispute matters to publishers, writers and other creators, as well as AI developers deciding how to obtain training material. It also raises questions about how copyright rules apply to a growing technology and an industry that contributes more than $1 trillion to the US economy.