SEO News: July 2026

Cloudflare Is Changing Its Rules for AI Bots: Watch Out for Googlebot!

Cloudflare is updating the way it identifies and blocks AI crawlers, which, starting September 15, may result in Googlebot being blocked on websites that prevent the training of artificial intelligence systems.

• New classification: Robots are now categorized by behavior into the Search, Agent, and Training categories.
• Multi-purpose crawlers: Robots such as Googlebot will be blocked if a page blocks one of their functions.
• Increase in AI requests: AI training now accounts for the majority of crawler requests, up from 20% in the spring of 2025.

For websites that use Cloudflare, this means that blocking AI training may unintentionally affect Googlebot's crawling and reduce visibility in search results.

Google Search Console Expands Monitoring to Include Social and Video Content

Google has announced the introduction of “platform properties,” a new type of property in Search Console. This new feature is designed to help content creators track the performance of their content from social media and video platforms in Google Search and Discover.

• Monitoring: Lets you see how posts from Instagram, TikTok, X, and YouTube appear in Google Search.
• Reports: Performance, Statistics, and Achievements reports are available for data analysis.
• Supported platforms: The new feature includes Instagram, TikTok, X, and YouTube.

For content creators, this means a better understanding of how their content from social media and video platforms performs in search results.

Google Ignores Cloudflare's Content Signals Guidelines, Mueller Confirmed

John Mueller of Google stated that the new Content Signals directives for `robots.txt`, introduced by Cloudflare, have no effect on crawlers or LLMs.

These directives only serve to "bloat" the `robots.txt` file and make it more difficult to maintain in the future.

• Google's stance: Google does not use `llms.txt`, `llms-author.txt`, or the `content-signal` directive in `robots.txt`.
• Wider use: Mueller is not aware of any other crawler or LLM that would follow these guidelines.
  

For SEO practitioners, this means that investing in the implementation of these guidelines with the aim of influencing Google crawlers or LLMs is currently ineffective.

Microsoft's Clarity Introduces Topic Insights for AI Visibility

Microsoft Clarity has introduced the Topic Insights feature, which provides an overview of how website content is cited in AI search results and generated responses.

• Action Recommendations: The tool helps identify gaps in content and suggests actions to improve AI visibility.
• Competitive Analysis: Lets you see which domains are cited alongside yours and where your competitors outperform you.
• Availability: This feature is available free of charge to all Microsoft Clarity users.

For content creators, this means an opportunity to respond to AI-driven demand before it becomes apparent in traditional search.

Merchant Center Now Includes Product Performance Overviews in AI Overviews

Google Merchant Center is introducing a beta performance report called the AI Performance Report. It shows how products are performing in AI mode and in AI Overviews.

• Pilot Phase: The report is currently available for select Merchant Center accounts in the U.S.
• Key Metrics: Provides data on share of voice, competitors’ average share, and query frequency.
• Data Type: Includes only organic AI traffic from free product listings.

The ability to monitor performance in AI Overviews thus makes it possible to optimize product data and strategies more precisely.

Google AI Overviews Now Generate Images Directly in Search Results

Google announced that its AI Overviews will now allow users to generate images directly within search results.

• Generation: Users will be able to create images from text prompts directly within AI Overviews.
• Models: The feature is powered by the Imagen 3 models and a new open-source image generation model.
• Security: Generated images will bear the SynthID watermark and will include security measures.

This feature signals a shift in how visual content is consumed, as it can now take place directly within search results.

Source: https://searchengineland.com/google-ai-overviews-now-lets-you-create-image-482163

EU AI Act in Effect: Mandatory Labeling of AI Content and Interactions

The EU AI Act, which introduces a requirement to transparently label AI-generated content and communications, takes effect on August 2. The Czech Artificial Intelligence Association also clarifies that the rules do not apply across the board to all AI output, but rather to five specific scenarios.

• Communication with AI: The user must be informed that they are communicating with an AI system (chatbot or voicebot).
• Technical Labeling: Providers must ensure that AI-generated text, images, audio, and video are labeled in a machine-readable format.
• Deepfakes and the Public Interest: Publishers must label AI-generated content that resembles real events or people, or that informs the public.

Publishers and content creators are therefore directly responsible for transparently labeling AI-generated material in order to avoid heavy fines and ensure compliance with the law.

Reddit: A Key Platform for SEO and Visibility in AI Search

A new analysis from Ahrefs highlights the dramatic rise in Reddit's importance for SEO and AI search; the platform has become the second-most cited domain for AI-generated answers.

• License Agreements: Reddit has license agreements with Google (US$60 million per year) and OpenAI.
• Impact on AI: The total value of contracts with AI platforms amounts to $203 million.
• Organic Search: In the U.S., Reddit ranks second with 727.3 million SEO visits per month.

Given these trends, maintaining an authentic presence and actively participating on Reddit are becoming an essential part of modern SEO strategies and visibility in AI search.

Google is cracking down on massive amounts of AI-generated content and imposing penalties for it

Many companies have tried to use generative AI for mass content production, but these initiatives often fail and trigger manual penalties from Google.

• Limited Resources: Google allocates resources for crawling and indexing based on a website's usefulness and authority.
• Freshness Factor: New content gets a temporary boost, but it needs genuine user signals to maintain that boost.
• Manual Penalties: Penalties for "Scaled Content Abuse" involving unsupervised bulk AI-generated content are rising sharply.

For publishers, this means that success with AI-generated content depends on its actual informational value, not just on quantity or checking off a checklist.

World Cup: Google Search Sees Record Usage

Nick Fox, senior vice president of Knowledge & Information at Google, and Robby Stein from the Search product team announced that on July 7, Google Search recorded its highest-ever usage. This occurred on the day Argentina defeated Egypt in the World Cup.

• No figures: Google did not provide specific figures or the methodology used to achieve the record.
• Previous records: Google mentioned records in Q1 2026 and mid-2022, also without providing specific figures.
• Methodology: It is not specified how “usage” is calculated, nor whether it excludes bots.

For publishers, this means that although Google remains the primary source for live events, increased usage does not necessarily translate to higher CTRs and clicks to their sites.

Google Clarifies Rules for Structured Data on Q&A Pages

Google's official documentation has been updated to clarify how to properly use Q&A (question-and-answer) structured data to display rich results.

• User Answers: The `QAPage` markup may only be used where users can add their answers to a question.
• One Question per Page: This type of structured data is intended for pages focused on a single question and its answers.
• Restrictions on FAQs and Blogs: Not suitable for FAQs generated by the website itself, how-to guides, or blog posts containing answers.

Proper implementation of Q&A structured data can thus significantly boost the visibility of pages containing user-generated questions and answers.

Claude Opus 5: A New Standard for Coding and Knowledge Work

Anthropic has introduced a new language model, Opus 5, which sets a new "state-of-the-art" benchmark in coding and knowledge-based tasks. The model is available for all paid plans and the Claude API.

• Availability: Opus 5 is included in all paid plans and the Claude API.
• Price: The price remains the same as for the previous version, Opus 4.8.
• Speed: The model is also available in Fast mode, which is 2.5 times faster.

The improved quality of AI models represents a leap forward in the possibilities for automating content creation and data analysis—not just for SEO specialists.

Sources

• https://www.searchenginejournal.com/cloudflares-ai-crawler-rules-can-block-googlebot/581385/
• https://developers.google.com/search/blog/2026/07/search-console-social-video-platforms
• https://www.seroundtable.com/google-cloudflare-content-signals-41631.html
• https://clarity.microsoft.com/blog/topic-insights-announcement/
• https://www.seroundtable.com/google-merchant-center-ai-performance-insights-41675.html
• https://searchengineland.com/google-ai-overviews-now-lets-you-create-image-482163
• https://www.lupa.cz/aktuality/od-2-srpna-nastava-povinne-oznacovani-umele-inteligence-prizpusobit-se-musi-rada-firem/
• https://ahrefs.com/blog/reddit-seo/
• https://www.searchenginejournal.com/scaled-ai-content-often-fails-googles-crawl-economics-explain-why/581325/
• https://www.searchenginejournal.com/google-says-search-hit-all-time-usage-high-during-world-cup/581796/
• https://developers.google.com/search/docs/appearance/structured-data/qapage
• https://x.com/claudeai/status/2080699497064083942?s=20