By Cliff Potts, CSO, and Editor-in-Chief of WPS News

Baybay City, Leyte, Philippines — April 21, 2026

The Double Standard Nobody Wants to Admit

For years, the United States and its allies have criticized other regions—particularly China and parts of Asia—for weak enforcement of intellectual property rights.

The argument is familiar:

  • Copyright violations
  • Unauthorized copying
  • Lack of enforcement

These practices are labeled clearly and repeatedly:

Piracy.

But when similar behavior appears inside Western technology systems, the language changes.

It becomes:

  • Innovation
  • Training data
  • Aggregation
  • Platform optimization

The behavior does not change.

Only the description does.

What Big Data Is Actually Doing

Modern AI and data platforms operate by ingesting large volumes of human-created content.

That includes:

  • Articles
  • Essays
  • Books
  • Artwork
  • Photography

This material is then:

  • Processed
  • Analyzed
  • Reassembled into outputs

Those outputs are monetized through:

  • Advertising
  • Subscriptions
  • Platform dominance

In many cases, the original creators:

  • Are not asked for permission
  • Are not compensated
  • Are not even aware their work is being used

That is the functional reality.

Why This Fits the Definition of Piracy

Traditionally, piracy has meant:

The use or reproduction of copyrighted material without permission or compensation.

The current system does not always reproduce content verbatim.

But it does:

  • Extract value from it
  • Depend on it
  • Generate revenue from it

The distinction between copying and extracting becomes less meaningful when the outcome is the same:

  • The creator’s work drives value
  • The creator does not share in that value

Whether the term used is “training” or “processing,” the economic effect mirrors what has historically been called piracy.

The China Comparison

Western governments frequently point to China as an example of systemic intellectual property abuse.

And in many cases, those criticisms have been valid.

But that raises a question:

Why is one form of unauthorized use treated as unacceptable, while another is normalized?

If:

  • Copying a film without permission is piracy

Then:

  • Using written, visual, or intellectual work to power commercial systems without compensation raises the same concerns

The inconsistency is difficult to ignore.

The Language Shield

Part of the reason this continues is language.

Terms like:

  • “Machine learning”
  • “Training data”
  • “Model development”

Create distance from what is happening.

They make the process sound technical and abstract.

But behind that language is a simple dynamic:

  • Human-created work is being used to generate value
  • Without direct compensation to the people who created it

Changing the vocabulary does not change the structure.

Why This Matters Now

This issue is becoming more urgent as AI systems expand.

The more these systems rely on:

  • High-quality writing
  • Original reporting
  • Creative work

The more they depend on the continued existence of creators.

If those creators are not supported:

  • Output quality declines
  • Original work becomes less sustainable
  • The system weakens over time

This is not just a fairness issue.

It is a structural one.

The Likely Outcomes

There are only a few ways this resolves:

  • Legal action defining limits on data use
  • Licensing systems for training and summarization
  • Revenue-sharing models between platforms and creators
  • Or continued extraction until the supply of high-quality input declines

None of these paths avoid the core issue.

They only determine how it is addressed.

The Bottom Line

The debate is not about whether technology should advance.

It is about whether the people whose work fuels that advancement are recognized and compensated.

When value is taken without compensation, the term “piracy” has historically been used.

If the same outcome is occurring under different language, the question is not whether the term is uncomfortable.

The question is whether it applies.

If you read this and it matters, help me keep it going: https://www.patreon.com/cw/WPSNews

References

Anderson, C. W., Bell, E., & Shirky, C. (2015). Post-industrial journalism: Adapting to the present. Columbia Journalism School.

OpenAI. (2023). GPT and the future of content generation.

Google. (2023). Search Generative Experience (SGE) overview. https://blog.google/products/search/generative-ai-search

World Intellectual Property Organization. (2021). Understanding copyright and related rights. https://www.wipo.int

Discover more from WPS News

Subscribe to get the latest posts sent to your email.