DuckDuckGo makes its 'no-AI' search engine easier to access as its traffic booms

Original Article ↗ Hacker News Discussion ↗

DuckDuckGo is drawing new interest by making its “no‑AI” search mode easier to access just as Google leans harder into AI-generated answers. Commenters weigh the trade-offs between AI summaries and traditional “ten blue links,” with many wanting explicit control over when AI is used, citing concerns about accuracy, manipulation, and the flood of AI‑generated “slop” in search results. The thread also compares DuckDuckGo’s quality and business model with alternatives like Google, Kagi, and Brave, highlighting a broader appetite for privacy‑respecting, less enshittified search—even if results are sometimes weaker than Google’s.

Overall response to DuckDuckGo’s “no-AI” search

  • Many welcome a clearly separated “no-AI” mode and say they just want classic “ten blue links” when doing web search.
  • Several long-time DDG users report it’s been their default for years, “good enough” or better than Google for general and English-language queries.
  • Others who recently switched say DDG still feels noticeably worse than Google or Kagi, especially for work, non‑English, or very specific technical/local searches.
  • Some appreciate DDG’s respect for search operators (e.g., inurl:, site:), which they feel Google increasingly ignores.

AI in search: demand, design, and backlash

  • Common view: users want to choose explicitly between a chatbot and a search engine; forced AI summaries in search are resented.
  • Several disable AI in DDG and other products entirely; they go to dedicated chatbots (ChatGPT, Claude, etc.) when they want AI.
  • A minority say Google’s AI answers are genuinely useful and sometimes better than its degraded regular search; others argue Google may have “nerfed” regular search to make AI look better.
  • Some praise Brave’s and Kagi’s approaches, where AI is optional, query-triggered, or gated by a special syntax.

Accuracy, safety, and manipulation concerns

  • Multiple stories of AI search hallucinating, including medical advice that could have led to unnecessary emergency care.
  • Users distinguish between “linking to an incorrect source” vs. the search engine itself stating wrong information as authoritative.
  • Worries that AI summaries will be trained increasingly on AI-generated “slop,” amplifying errors.
  • Strong suspicion that AI answers will be used to blend ads invisibly into “neutral” summaries.

Search quality, spam, and indexing issues

  • Broad frustration with modern web search: SEO spam, AI-generated pages, paywalls, popups, and Reddit’s preferential deal with Google.
  • Desire for engines that aggressively filter AI/SEO junk and surface older, higher-signal content.
  • DDG is criticized for poor non-English results and for technical limits (e.g., ~200 results cap, 10 results/page, buggy advanced syntax, JS challenges on “lite” search).

Economics and politics of “no-AI”

  • Some see “AI-free” products as a new premium niche: mass market gets cheap AI output; paying users get human-curated, privacy‑respecting services (e.g., Kagi, DDG).
  • Others are skeptical, viewing coverage and growth stats (like “+30% traffic”) as marketing without meaningful absolute numbers.
  • Thread also spirals into debate about AI displacing jobs, whether technologists should resist or embrace it, and calls for collective action (e.g., unionization).