Welcome to our monthly SEO news roundup for August 2026. In this edition, we dive into critical updates affecting your search visibility and technical SEO strategy, including new protocol developments from Cloudflare, important Google and Bing updates, and the evolving landscape of AI-driven visibility.
Cloudflare launches PACT protocol for detecting human users
Cloudflare is developing the new open-source standard “Private Access Control Tokens” (PACT) together with Google Chrome, Microsoft Edge, Mozilla Firefox, and Shopify. The goal is to verify genuine human interactions and authorized bots in a data-compliant manner without CAPTCHAs via anonymous validation. For SEO and GAIO, this development marks the beginning of a clear separation at the infrastructure level between human identity and the authorization of autonomous AI agents. Will we soon never have to click on traffic lights and zebra crossings again? Read more here.
Cloudflare rule adjustment can unintentionally block Googlebot
Cloudflare will strictly divide AI crawlers into the three categories “Search”, “Agent”, and “Training” in the future and will apply the strictest defined rule set for combined bots starting September 15th. This creates a significant SEO risk for companies in the European region, as a blanket blocking of AI training bots at the network level can also completely lock multi-purpose crawlers like Googlebot or Bingbot out of the web server. Read more here.
Google recommends 304 status code for better crawl budget
Google has updated its documentation on crawl budget and explicitly recommends the use of the HTTP status code 304 (Not Modified) for unchanged pages. If configured correctly, this should also reduce the load on our servers when bots crawl less. Perhaps in part a compensation for the increased visit frequency of AI bots and scrapers. In principle, server log file analysis, crawl budget optimization, and performance optimizations remain a regular fundamental duty for Tech-SEOs. Read more here.
Google tests opt-out for generative AI features in Search Console
Website operators can specifically exclude their content from the AI features of Google Search via a new Search Console setting without losing their regular ranking in organic search results. For the European market, however, the decision to opt out is currently a blind flight, as Google does not yet provide click data for AI Overviews and there is uncertainty about whether classic placements like Top Stories within the AI elements will also be lost. Read more here.
Bing to discontinue SOAP/POX APIs on August 31, 2026
Anyone who has built SEO tools and reporting dashboards must now check the connection to Bing Webmaster Tools, or data might stop flowing starting in September. Microsoft is permanently retiring the deprecated SOAP and POX interfaces of the Bing Webmaster Tools on August 31, 2026, and is urging users to migrate to the REST/JSON API.
Notice: The SOAP and POX/HTTP APIs will be retired on August 31, 2026. After this date, requests to these endpoints will no longer be served. Migrate to the JSON/HTTP (REST) API before August 31, 2026 to avoid service interruption.
Urgent need for action, therefore, especially for all of us who operate automated indexing pipelines and sitemap management tools. Read more here.
Common Crawl as an upstream bottleneck silently decides on the AI visibility of brands
Visibility in LLM response systems is shifting from classic indexing to upstream pre-training pipelines, the basis of which consists of over 60% Common Crawl data sets (Mozilla Study 2024). While brands are optimizing their on-site GAIO strategies, CDN default settings like those of Cloudflare or aggressive WAF rules block the CCBot at the infrastructure level silently and without manual intervention by the site owners. Those who are not represented in the monthly crawl systematically lose the chance to be anchored as factual ground truth in the model knowledge of future models from GPT, Claude, or LLaMA. Infrastructure providers and CDN platforms thus become involuntary gatekeepers of AI presence, which pushes first-party content out of the training data without professional tech auditing. The winners of this shift are agile players with clean server architecture and high harmonic centrality, while established brands are stealthily losing digital market presence due to standard technical blocks.
Strategic Takeaway: AI visibility requires expanding Technical SEO to include the infrastructure layer: Enabling and continuously monitoring training crawlers at the CDN and edge level is the fundamental prerequisite for any downstream Generative Engine Optimization.
If you are unsure whether you are utilizing all opportunities and know and control all risks, speak with us today.
Bonus: Common Crawl has now published a compressed guide for ensuring basic AI visibility: Check it out here.
