Copyright Regulation Of AI Narrative Generation In Poland.

1. Introduction: AI Narrative Generation and Copyright in Poland

Artificial Intelligence systems can generate stories, scripts, articles, and narratives automatically. These outputs raise important copyright questions such as:

Who is the author of AI-generated narratives?

Can AI-generated text receive copyright protection?

What happens if AI systems use copyrighted material during training?

How do courts treat AI-assisted storytelling tools?

In Poland, these issues are addressed primarily under:

Polish Copyright and Related Rights Act

Directive (EU) 2019/790 on Copyright in the Digital Single Market

Poland follows the European copyright tradition, which requires that a protected work must be an original intellectual creation of a human author. AI cannot legally be recognized as an author.

Therefore:

Pure AI-generated narratives cannot receive copyright protection.

Narratives created with meaningful human creative input may be protected.

AI developers or users may hold rights only if human creativity is demonstrable.

2. Legal Principles Governing AI-Generated Narratives in Poland

A. Requirement of Human Authorship

Under the Polish Copyright and Related Rights Act:

A “work” must be:

Original

Creative

Created by a human author

Therefore:

ScenarioCopyright Status
Human writes narrative using AI suggestionsProtected
AI generates narrative autonomouslyNot protected
Human edits AI narrative creativelyProtected for edited portions

B. AI as a Creative Tool

Polish scholars and courts generally treat AI as similar to software tools like Photoshop or word processors.

Thus:

AI assists creativity.

The human directing the AI remains the author.

C. Text and Data Mining (TDM)

AI narrative systems train on large datasets of text. The EU directive implemented in Poland allows text and data mining under certain conditions:

For research institutions

When rights holders have not opted out

Provided the use does not substitute the original work

This has implications when AI models generate narratives based on training data.

3. Case Laws Relevant to AI Narrative Copyright

Although Poland has only recently begun addressing AI directly, courts rely on existing copyright principles. The following important Polish and European cases shape how AI narrative generation is interpreted.

Case 1

Infopaq International A/S v Danske Dagblades Forening

Background

A Danish media monitoring company used automated software to scan newspaper articles and generate short summaries.

The system:

Digitized newspapers

Extracted 11-word snippets

Generated automated text reports

Legal Issue

Whether small fragments of text generated automatically could still be protected by copyright.

Court Decision

The court ruled that:

Even very small text segments can be protected if they contain original expression.

Automated extraction still involves copyrighted material.

Importance for AI Narrative Generation

This decision influences Poland because EU law applies across member states.

Implications:

AI-generated narratives may infringe copyright if they reproduce protected fragments.

Even small textual similarities may lead to infringement.

AI narrative tools must avoid copying original expression.

Case 2

Football Dataco Ltd v Yahoo! UK Ltd

Background

This case involved football fixture lists created partly through automated processes.

The question was whether automatically generated data structures could receive copyright protection.

Court Decision

The court held that:

Works must reflect author’s own intellectual creation.

Purely mechanical or automated production does not qualify.

Significance for AI Narratives

This principle directly affects AI storytelling.

If a narrative is produced entirely by AI:

There is no human intellectual creation

Therefore no copyright protection

However, if a human:

designs prompts

edits the narrative

structures the story

then protection may exist.

Case 3

Deckmyn v Vandersteen

Background

A political party created a parody based on a famous comic book cover.

Issue

Whether using elements of an existing creative work in new content qualifies as parody or infringement.

Decision

The court ruled that parody must:

Evoke the original work

Be noticeably different

Express humor or critique

Relevance to AI Narratives

AI systems sometimes generate narratives in the style of famous authors.

This case suggests that:

Generating stories closely imitating a protected work could violate copyright.

AI narratives must be transformative rather than derivative.

Case 4

SAS Institute Inc v World Programming Ltd

Background

A company recreated software functionality based on observing another program.

Legal Question

Whether functionality, programming language, or system behavior could be copyrighted.

Court Decision

The court held that:

Ideas and functionality are not protected

Only original expression is protected

Impact on AI Narrative Systems

AI narrative engines often replicate:

writing structures

storytelling patterns

narrative styles

According to this principle:

Ideas and narrative structures are free to use

Only specific textual expressions are protected.

Case 5

Painer v Standard Verlags GmbH

Background

A portrait photographer sued newspapers that reproduced her photo.

Issue

Whether photographs contain sufficient creative choices for copyright.

Decision

The court recognized copyright because the photographer made choices such as:

lighting

angle

composition

Importance for AI Narratives

This case establishes a broader principle:

Copyright exists when creative decisions are made by a human.

Thus:

If a human guides AI storytelling through prompts and editing, those choices may create copyright.

Case 6

Pelham GmbH v Hütter

Background

A music producer sampled a small portion of another recording.

Issue

Whether very small fragments could be used freely.

Decision

The court ruled that recognizable copying of sound fragments requires permission.

Relevance to AI Narratives

AI narrative systems trained on books may reproduce recognizable passages.

This case implies:

Even short textual similarities may infringe copyright.

4. Practical Legal Challenges in Poland

1. Ownership of AI Narratives

Possible claimants include:

AI developer

user who prompts the system

organization publishing the story

In Poland, courts would likely recognize the human who contributed creative choices.

2. Training Data Liability

If AI was trained on copyrighted books without permission:

rights holders could claim unauthorized reproduction during training.

This debate is ongoing across Europe.

3. Plagiarism and Similarity Detection

AI narratives may unintentionally replicate phrases from training data.

Polish copyright law protects original expression, so substantial similarity could lead to infringement claims.

5. Future Legal Developments

Poland is expected to adapt its copyright framework in response to EU initiatives such as:

AI governance rules

transparency obligations for generative models

training data disclosure requirements

Future regulation may require AI systems to:

disclose training sources

prevent copyrighted text reproduction

track AI-generated content.

6. Conclusion

Copyright regulation of AI narrative generation in Poland is based on three core principles:

Human authorship is essential for copyright protection.

Pure AI-generated narratives are not protected works.

Copyright infringement may occur if AI reproduces protected text from training data.

The discussed cases—including
Infopaq International A/S v Danske Dagblades Forening,
Football Dataco Ltd v Yahoo! UK Ltd,
Deckmyn v Vandersteen,
SAS Institute Inc v World Programming Ltd,
Painer v Standard Verlags GmbH, and
Pelham GmbH v Hütter

provide the legal foundation used by Polish courts when addressing AI-generated narratives.

LEAVE A COMMENT