# Could AI detection eventually become a fundamentally unsolvable problem, similar to the halting problem? [D] — retained record

> Retained reddit record from reddit:MachineLearning.

Canonical: https://highsignal.app/data/records/ba3b9eea2f09bcf0

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:MachineLearning)  published 2026-10-28 16:42 ingested 2026-10-29 02:38

[original source ↗](https://www.reddit.com/r/MachineLearning/comments/1xhtabq/could_ai_detection_eventually_become_a/)

## retained text



```
In theoretical computer science, we have the halting problem, which proves that certain problems cannot be solved by any general algorithm.  
This made me wonder about something similar in machine learning.  
Today, we can often distinguish AI-generated content from human-created content, at least statistically. But as AI models improve, could we eventually reach a point where the two become completely indistinguishable?  
Not just because our detectors aren’t advanced enough, but because there is no longer enough information in the content itself to tell them apart, regardless of how powerful the detector is.  
In other words, could AI detection eventually hit a fundamental mathematical limit rather than just a technological one?  
I know this wouldn’t be the same kind of undecidability as the halting problem, but I’m curious whether there’s an established theoretical framework for this.  
What do you think?
```



document fetched 2026-10-29 00:28 · canonical https://reddit.com/r/MachineLearning/comments/1xhtabq/could_ai_detection_eventually_become_a
