Google's AI answers reducing search traffic by 40%
Google’s new AI-generated search answers are reportedly cutting click-throughs to external sites by around 40%, while leaving ad clicks largely intact. Commenters welcome faster, cleaner answers after years of SEO spam and ad-choked pages, but worry this shift will starve independent publishers of traffic and incentives to create new content, accelerating the decline of the open web. Many predict a future where AI agents, not browsers, mediate most information access—raising fresh concerns about accuracy, bias, monetization, and how original work will be funded.
Impact on traffic and business models
- Multiple site owners report steep organic search declines (often ~80%), especially for content/SEO-driven sites; some pivot to direct sales or outbound outreach.
- Others see fewer but better-qualified leads when users arrive via AI assistants that have pre-filtered options.
- Concern that if search engines stop sending traffic, independent sites will shut down, undermining the very corpus AI depends on.
User experience & search quality
- Many users like AI answers for simple factual queries (“who’s hosting”, “when is X”), avoiding clickbait, cookie banners, pop‑ups, and long SEO intros.
- Others say AI overviews are often incomplete, wrong, or outdated; they prefer “actual information” from source sites or use AI only with citations.
- Several argue Google search quality had already collapsed due to ads and SEO; AI is a band‑aid, not a cure.
SEO slop, AI slop, and web decay
- Strong consensus that traditional SEO content and review blogs heavily degraded the web.
- Now AI-generated “slop” is rapidly filling results; some say even Kagi and YouTube are getting flooded.
- Some see AI as a partial antidote to SEO farms; others warn it just creates a new layer of low‑quality synthesis on top of low‑quality inputs.
Content creators, incentives, and ethics
- Widespread fear that summarization without clicks removes incentives to create: blogs, guides, niche history sites, forums, and even Wikipedia editing.
- Debate whether this is “theft” or legitimate reuse of uncopyrightable facts; some frame it as enclosure/value capture rather than literal stealing.
- Suggested fixes include revenue sharing, licensing standards, content gating, micropayments, or poisoning AI scrapers; feasibility is unclear.
Alternatives & user adaptation
- Many report heavy migration to ChatGPT/Claude for “search,” plus Kagi, DuckDuckGo, Brave, etc. for link-based queries.
- Niche communities increasingly live in walled gardens (Discord, Telegram, forums), partially insulated from AI scraping but harder to discover.
Future of ads, SEO, and AI search
- Expectation that “AI SEO” (GEO/AEO) and paid influence over model outputs will emerge, along with ads embedded in AI answers (already seen in some products).
- Some foresee ad-free, local, user-controlled agents; others argue thin clients, closed models, and profit motives make deeper enshittification likely.