Why Does AI Sometimes Duplicate Content as Raw Code Instead of Readable Format?

I asked an AI tool to duplicate a page… and it recreated it as raw XML/code.

What happened

The request was straightforward:

• duplicate a page

• remove certain elements

• rename it

The AI:

• created the new item with the correct title

• but the content appeared as raw structured code instead of readable text

Why this happens

Short Answer: The AI pulled the system’s storage format, not the rendered view.

Most modern platforms store content in two forms:

• Rendered view → what users see in the UI

• Storage format → structured data behind the scenes (often JSON, XML, or similar)

In this case: The AI accessed the underlying structured format and reproduced it directly—without converting it back into a human-readable version.

Why it feels inconsistent

Because it is.

AI systems don’t “copy” content the way native system features do. They:

• retrieve content

• interpret it

• regenerate it

During that process, they may:

• use the rendered version (looks correct)

• use the storage format (looks like code)

There’s usually no visible control over which version is used.

What you can do today

If this happens:

1. Simplify the request: Multi-step instructions increase the chance of format issues.

2. Break tasks into steps: first duplicate or extract content > then modify it

3. Provide explicit content: Instead of referencing an entire page, specify what should be copied.

If it keeps happening

Capture the timestamp, source location and prompt used then report it to the platform provider. This helps identify:

• which format was retrieved

• where conversion failed

Takeaway

AI doesn’t duplicate content the way systems do natively. It:

• reads content

• reinterprets it

• rebuilds it

That flexibility is powerful—but introduces variability.

When to use native tools instead

If you need:

• exact formatting

• consistent structure

• predictable results

Use the system’s built-in duplication or copy feature.

Use AI when:

• transforming content

• summarizing

• restructuring information

  
  

AI is great at recreating meaning.

Systems are still better at preserving structure.