A Silent Book With a Loud Message

What does a book with no words say about the future of creativity?

At the 2026 London Book Fair, thousands of authors released a striking protest: an “empty” book titled Don’t Steal This Book. Instead of stories or essays, the pages contain only the names of roughly 10,000 writers, symbolizing works they say artificial intelligence companies are using without permission.

The unusual publication highlights growing tension between the publishing industry and the rapidly expanding AI sector. As generative AI systems rely on massive datasets that often include books, articles, and creative works, many writers are asking a fundamental question: Who owns the words that train AI?

Why Authors Created an Empty Book Protest

The empty book protest against AI copyright was organized by composer and AI ethics advocate Ed Newton-Rex. Its goal was simple but powerful: show the scale of creative work potentially used to train AI models without consent.

Notable authors including Kazuo Ishiguro and Philippa Gregory joined the protest.

Their concern centers on how generative AI systems learn. Large language models are trained on enormous collections of text, which may include copyrighted material. Critics argue that this process allows companies to benefit from authors’ work without compensation.

The empty book acts as a symbolic ledger of creators demanding recognition and control over their intellectual property.

The AI Training Data Controversy

The debate about AI copyright and training data has intensified over the past two years.

AI developers argue that training models on large datasets can fall under “fair use,” particularly when the system does not reproduce original works directly. However, many creators disagree. They claim the practice effectively turns their writing into raw material for AI-generated content.

Recent reports reveal the scale of the issue. Some AI companies have scanned millions of books or used large shadow libraries containing copyrighted works to build training datasets.

At the same time, publishers and authors have filed lawsuits against technology companies over alleged copyright infringement and data scraping practices.

These legal battles may ultimately determine how AI development interacts with copyright law.

The Policy Debate Around AI Copyright

The empty book protest against AI copyright arrives at a crucial moment for policymakers.

In the United Kingdom, the government is evaluating changes to copyright law that could allow AI firms to use copyrighted works for training unless creators opt out.

Critics say this flips the traditional copyright model. Instead of companies asking permission, authors would need to actively block AI access to their work.

Supporters argue that easier data access could accelerate innovation in AI research and maintain global competitiveness.

For governments, the challenge is balancing two powerful forces: technological progress and protection of human creativity.

What This Means for the Future of AI and Publishing

The empty book protest highlights a broader transformation in how knowledge and creativity intersect with artificial intelligence.

If AI developers must license training data, it could create new revenue streams for writers and publishers. Some industry groups are already exploring licensing frameworks to manage AI access to books.

If courts rule that training qualifies as fair use, AI companies may continue expanding datasets with minimal restrictions.

Either way, the protest underscores a key reality: the AI revolution depends heavily on human-created knowledge.

The empty pages in Don’t Steal This Book may contain no words, but they carry a powerful message about the future of authorship in the age of AI.

Fast Facts: Empty Book Protest Against AI Copyright Explained

Why did authors publish an empty book to protest AI copyright?

The empty book protest against AI copyright highlights how writers believe their work is used to train AI systems without permission or payment. The book symbolically lists thousands of authors whose works may have contributed to AI datasets.

How does AI training relate to copyright law?

The empty book protest against AI copyright focuses on AI models trained on large datasets that may include copyrighted books. Companies often argue this qualifies as fair use, while authors say it violates intellectual property rights.

What could change after the empty book protest?

The empty book protest against AI copyright may influence future regulation. Governments and courts are considering whether AI companies must license training data or compensate authors for the use of their copyrighted works.