How often do AI overviews show lower than position 1 in Google's search results? I've seen many posts recently that AIOs are appearing in lower positions, but it's helpful to understand the frequency. Recently, I discovered that you can get the AIO position data by device for my clients when *exporting* from Semrush (a position isn't assigned within the dashboard itself). If we were to conduct this same analysis using Wikipedia, being the most frequently appearing domain in AIOs with millions of query ranking data, I can see the frequency of ranking position. Here are some key takeaways: AI Overviews - Desktop - US Position 1 Occurrences: 5,713,219 (99.6%) Position 2 Occurrences: 14,713 (0.26%) Position 3 Occurrences: 5,969 (0.10%) Position 4 Occurrences: 1,916 (0.03%) Position 5 Occurrences: 401 (0.007%) AI Overviews - Mobile - US Position 1 Occurrences: 1,501,503 (99.3%) Position 2 Occurrences: 6,154 (0.41%) Position 3 Occurrences: 2,733 (0.18%) Position 4 Occurrences: 1,190 (0.08) Position 5 Occurrences: 462 (0.03%) Other Insights • The lowest position recorded for an AIO on desktop was position 99. This is due to the AIO appearing in the righthand panel, which is considered the bottom of the SERP from a ranking perspective. • AIOs can effectively appear in any position on the page, but position 5 tends to be the lowest position for most sites. • While AIOs can appear lower than position 1, it is still quite rare based on the Semrush dataset, with the chances of appearing lower than 1 being less than 1%. Note 1: Semrush position tracking for AIOs likely works different to GSC, in that GSC will only record a position if the link is visible. For Semrush, they go a step further and record a ranking if the favicon for the site appears in the citations – this is what signifies if the AIO shows the link to a page once expanded. Note 2: there are instances where AIOs will have links that appear as underlined within the AIO (rather than as a separate citation. In this instance, it is more likely that Semrush and GSC would have a more similar recording of the data from a position standpoint.

AI Overviews are stealing your traffic in 2025, but SEO Stuff ( has already turned it into a sales channel. Let me show you how in this post. Mind you, AIOs aren't just a feature, they're a new search layer dominating above-the-fold. Oh, by the way, if you want some free advice on how to surface your site + some data from the SEO Stuff dashboard, RT this + follow me + reply with "SEO Stuff AI Overviews" and I'll DM you. Key insights from a recent study by Kevin Indig and Eric van Buskirk: - Users skim AIOs like fact sheets - Click out 30% less than before - 58% choose familiar brands over better answers The game has changed: You need brand recognition, not just relevance. Want the shortcut? Use SEO Stuff or more specifically the Gold Plan ( 80%+ customer return rate speaks for itself. Want to do it yourself? Here's the full roadmap. To win AIO clicks, focus on: - Getting mentioned in high-authority publications ( - Featured content on Reddit, YouTube, forums (Reddit marketing in monthly plans: - Creating unique, citation-worthy content (60 SEO/AIO optimized articles: AIOs cite brands with: - Backlinks from trusted sources (Premium ones: - Organic online mentions - Conversation presence Make your content AI-citable: - Structured answers with clear H2s/H3s - Original research and data (LLMs love this) - Plain language summaries - Regular updates for freshness Critical: Mobile UX is everything. Mobile users are the only ones still clicking AIOs. The Validation Factor: 30% of users leave AIOs to validate answers on Reddit, YouTube, and forums. If you're not there, you won't convert. You need two content types: 1. For AI citation (concise, factual, structured) 2. For post-AI clicks (experiential, case studies, skeptical-friendly) Your Action Plan 30 Days: - Audit top 25 keywords for AIO triggers - Optimize top 5 pages for mobile UX - Join key Reddit communities - Monitor brand mentions in AIOs 90 Days: - Publish AI-citable guides - Launch validation YouTube content - Build authority backlinks - Track mobile conversions 6 Months: - Release original research - Scale Reddit/YouTube presence - Create validation-first landing pages - Establish regular media features You're not optimizing for Google anymore. You're optimizing for Google AI, ChatGPT, Claude, Gemini, Perplexity, and Grok. They all prioritize authority, structure, and trust. Bottom line: The winners in 2025 aren't #1 in rankings, they're trusted everywhere users go to validate information.

近年来AIOS生态系统的发展确实令人瞩目，但其未来能否持续成功，仍需深入思考几个关键问题。 虽然Sh埃尔经过（Sheer Network）等去中心化存储方案的推出提升了AIOS的技术叙事，但存储赛道本身竞争激烈，已有Filecoin、Arweave等成熟项目占据市场。AIOS若想真正突围，不能仅靠技术堆砌，而应找到更精准的差异化定位，比如结合AI代理的智能数据索引或隐私计算能力，使其存储网络不仅仅是存储，而是成为AI训练和推理的高效数据层。 AIOS生态目前仍以复制以太坊项目为主，这固然能快速丰富生态，但也可能陷入跟随者困境。真正的创新应来自底层架构的突破，而非简单移植。例如AIOS主打AI代理操作系统，是否可以考虑优化智能代理的自治协作能力？或者利用零知识证明（ZKP）增强AI模型的可验证性？这些方向若能突破，才能真正形成护城河。 代币经济模型的设计也值得商榷。AIOS代币价格上涨固然可喜，但若缺乏实际应用场景支撑，仅靠市场情绪推动，其市值能否持续增长存疑。未来应更注重代币的实用性设计，比如质押代币以获取AI算力、治理投票权，甚至作为AI代理服务的支付媒介，让代币价值与生态发展深度绑定。 AIOS的长期竞争不仅在于技术，更在于生态协作。参考Linux和Android的成功经验，开源社区和开发者生态的培育至关重要。AIOS可借鉴国产操作系统如deepin的开源策略，降低开发门槛，吸引更多独立开发者和企业参与，避免陷入闭门造车的困境。 总的来说，AIOS目前的发展势头良好，但真正的挑战在于如何从“技术整合”转向“原创突破”，并在激烈的公链竞争中找准自己的独特定位。若能围绕AI代理构建一个开放、高效且可持续的生态系统，AIOS或许能成为下一代去中心化操作系统的有力竞争者。 #AIOS #DecentralizedOS #AIBlockchain #Layer1 #DePIN

Not all sites have to lose traffic to AIOs, here' a site we are working on that's been able to grow despite AIOs eating into informational AND commercial/transactional queries. It is still viable in 2025 to grow and scale, but the approach needs to be different, there needs to be a degree of "compensatory" content but with more effort per piece of content produced.

新しい転生先のホロリスの垢(@ribose_Monster)を作りました！これからこちらでも活動しますのでよろです！#転生 #新垢 #ホロライブ好きと繋がりたい #ホロリス

🚨 NEW GEM ALERT 💎 @aios_foundation $AIOS has joined CoinToRise! 🎉 🔗 Solana 💫 Project Highlights: ✨ Carefully Vetted 🔒 Security Verified 📈 Growth Potential 🔍 Explore: #NewGem #AIOS #CoinToRise #Solana

3. Similarly, when looking at the domian level, the average was a 16% overlap between AI Mode and AIOs.

🔎Google is now generating “AI Overviews” without marking them as AIOs. @rustybrick

しごおわ！仕事が辛くても推しのホロメンのイラスト見て疲れが癒されたよ…！🥰

working title: aiOS estimated completion: sooner than you think target platform: everything you browse 🧵 Thread incoming tomorrow

0gm legends 🙌 AI shouldn’t be locked in Big Tech’s basement. @0G_labs is building the future where you OWN your agents, your data, your intelligence. ✅ Fast ✅ Verifiable ✅ Decentralized This is how we take AI back. 🐼💜 #0GLabs #AIOS #Web3

オート育成、分かってきたようなよく分かんないような…… #ウマ娘

Google's AI Mode will eventually roll out in the UK and there's a real potential that it could be more disruptive than AIOs. Luckily we have some insights from @semrush who analysed 150K citations to see how AI tools compare to real Google results. 📊 Google AI Mode cited URLs from the top 10 just 35.41% of the time. Perplexity? 82.34%. AI is reshaping search but that doesn't mean it always surfaces the best content. Full study 👇 #SEO #AIMode #SemrushAmbassador