awesome-ai-radar

Noise Hall of Fame

The best explanations of why hyped things don't matter.

AI Search Engines Are Not Replacing Google (Yet)

What happened

Multiple outlets ran headlines this week claiming AI search engines (Perplexity, SearchGPT, Gemini Search) are "killing Google Search" based on a SimilarWeb report showing combined AI search traffic up 300% year-over-year. The actual numbers: AI search engines collectively handle approximately 2-3% of global search queries. Google's search market share dipped from 91% to 88% — with most of that loss going to TikTok and Reddit search, not AI tools.

Why it matters

This is a recurring pattern in AI coverage: impressive-sounding relative growth numbers (300% increase!) masking tiny absolute numbers (2-3% of queries). AI search tools are genuinely useful for certain queries — research synthesis, technical questions, comparison tasks — but the "Google is dead" narrative is at least 3-5 years premature. The real disruption to Google isn't from AI search replacements, it's from AI tools that eliminate the need to search at all (coding assistants, in-app AI features, agent workflows). Don't confuse the two narratives.

Who should pay attention

Founders building AI search products (your TAM is real but smaller than VCs think), content marketers worried about AI search disrupting SEO (it will, slowly, but Google is still the game), and anyone making strategic decisions based on "Google is dying" takes.

View Source →