AI Labs Avoid 'AI Slop' by Using Pre-2022 Books for Training

Search Engine Journal3h ago·1 min readStrategy & Trends

AI Summary

AI labs are actively sourcing pre-2022 books to train their models, deliberately avoiding newer content they deem 'AI slop.' This move, coupled with the development of text watermarking, signals a strategic shift away from content-at-scale strategies.

⚡ Marketer Insight

The discerning use of training data by AI labs highlights a growing concern about the quality and authenticity of AI-generated content, suggesting that future AI models may be judged by their foundational data, not just their output volume.

#ai training data#content quality#text watermarking

Original article

Search Engine Journal

Read full article →