The Pre-2022 Filter: How LLMs are Changing Book Trust and Consumption

The Rise of the 'Pre-2022' Trust Heuristic

Readers are increasingly adopting a subconscious or conscious preference for books and digital content published on or before 2022. This trend stems from the widespread integration of Large Language Models (LLMs) into content creation, leading to a perceived decline in the reliability, effort, and authenticity of post-2022 publications.

The Erosion of the Human Quality Signal

For many, the publication date serves as a proxy for human effort and rigorous editing. Pre-2022 works are viewed as having a higher probability of being manually typed, checked, and proofread, whereas newer works may be the product of automated generation.

The Value of Human Effort

One perspective is that the manual effort involved in traditional writing represents a "vote of quality." As one observer noted:

"Thinking, Writing, Editing, Proofing, Publishing, Distributing, Selling... At every step, a human had the opportunity to say 'no, this is bad.' And the fact that they didn't is a vote of quality and reputation."

The "Industrialization of Thinking"

There is a growing sentiment that AI-generated content lacks the intrinsic value of human thought, described by some as the "industrialization of thinking" or comparable to "frozen pasta from the shelves." This suggests that the process of creation is as important to the reader as the final output.

Market Saturation and the "AI Slop" Problem

The proliferation of AI tools has led to a surge of low-quality, AI-generated non-fiction, particularly on platforms like Amazon. This has created several tangible issues for consumers:

  • Low-Quality Reference Material: Users report that AI-generated reference books often lack fact-checking, proper editing, and professional layout, resulting in poor-quality technical guides.
  • High-Volume Publishing: Some "authors" are publishing an unrealistic volume of work—up to two books per week—which signals automated generation to discerning buyers.
  • Supply Chain Issues: There are reports of "AI-generated copy text" being sold under genuine-looking covers, where the interior content does not match the promised subject matter.

Impact Across Different Mediums

This distrust extends beyond physical books to various forms of digital information:

  • Technical Documentation: Some argue that tech documentation has seen a "race to the bottom," moving from comprehensive vendor manuals to sparse README files, with AI now filling the void as an individualized but narrowly focused "teacher of record."
  • Online Forums: Users are applying similar date filters to search results on Stack Overflow, Reddit, and Hacker News, favoring posts dated before 2023 for stable information.
  • Fan Fiction: Even in creative communities, readers are checking publication dates to avoid content that feels "bewitched" or lacks emotional resonance.

The Dilemma for Modern Human Authors

The shift toward pre-2022 filtering creates a paradoxical challenge for authors who write manually today. Because publication dates are updated upon revision, a human-authored book updated in 2026 may be unfairly dismissed as AI-generated.

Furthermore, the reliability of AI detection tools is questioned. One author reported that a long article written entirely by hand was flagged as 60% AI-generated by detection software, suggesting that humans may struggle to prove their authorship in a post-LLM world.

Counter-Perspectives and Future Outlook

Not all readers believe the 2022 cutoff is a sustainable or necessary strategy. Some arguments against this trend include:

  • The Role of Gatekeepers: Proponents of the traditional publishing model argue that publishing houses and taste-makers will continue to filter for quality, regardless of the tools used.
  • The Need for New Standards: Some suggest that instead of clinging to old dates, the industry must develop new standards for verifying human authorship and quality.
  • AI as a Tool for Research: There is a belief that AI will eventually evolve to provide better, heavily sourced, and verifiable facts, moving beyond the current "just-good-enough-to-be-dangerous" phase.

Sources