Claim
Production cost, not just algorithmic filtering, historically prevented search index spam at scale. LLMs collapsed that cost to near-zero, removing the quality gate that cost had imposed.
Mechanism
Search index quality was maintained by two forces together: Google's algorithmic signals and the real cost of producing content in bulk. Producing thousands of pages per day required capital, editorial staff, and operational overhead. That friction was a natural quality floor. Publishers below a certain quality-to-cost threshold could not sustain volume. Programmatic SEO lowered the floor. LLMs removed it. Any team can now generate plausible content at near-zero marginal cost per page. The only remaining check is Google's ability to detect and demote low-quality content algorithmically, without the production-cost barrier that used to limit supply.
Conditions
Holds when: content production is the bottleneck for index pollution. The same mechanism applies to AI-retrieval indexes that ingest the same source material used by traditional search.
Fails when: distribution is the bottleneck rather than production. Low-cost content harms index quality only where it achieves sufficient domain authority and distribution to be indexed and ranked. Sites without distribution cannot pollute the index at volume even at zero production cost.
Evidence
Ray connected the progression from programmatic SEO to LLMs as an escalating cost collapse, with index quality as the casualty.
"The marginal cost of publishing is one of the only things standing between a healthy index and total chaos (spammy search results)"
Signals
- Google algorithm updates target content categories showing sudden volume spikes without matching engagement signals.
- CTR drops site-wide despite stable rankings, as Google deprioritizes verticals flooded with generated content.
- Your production cost per page is higher than competitors, and your traffic holds after updates while theirs falls.
Counter-evidence
Google has systematically updated spam and Helpful Content signals to detect AI-generated content at scale. Domain authority, E-E-A-T signals, and engagement metrics partially substitute for the production-cost floor. AI-assisted editorial workflows can produce high-quality content at reduced cost, meaning low cost does not automatically mean low quality. The degree of index degradation varies sharply by vertical: high-trust verticals (medical, legal, financial) have additional algorithmic scrutiny that limits the damage.
Cross-references
- A Google penalty removes a page from every AI surface at once, not just Google search: index quality problems compound when low-quality content is also ingested by AI retrieval surfaces.
- A single seeded fake claim can self-confirm in AI Overviews: the reinforcing dynamic between low-cost content production, AI training data contamination, and further quality degradation.