←back to thread

646 points blendergeek | 2 comments | | HN request time: 0s | source
Show context
sedatk ◴[] No.42734326[source]
Both ChatGPT 4o and Claude 3.5 Sonnet can identify the generated page content as "random words".
replies(1): >>42736142 #
1. tlonny ◴[] No.42736142[source]
Given the size of the training data - I don’t think it would economical to validate all training data with high-end LLM models.
replies(1): >>42740871 #
2. sedatk ◴[] No.42740871[source]
True. Maybe it can be dumbed down to a low-end model specifically for this type of detection.