Search and AI Changes: August 31–September 6, 2026
Ten material search and AI changes from August 31–September 6, organized by the decisions publishers, SEOs and advertisers need to make.
Tagged collection
Latest articles tagged “AI Search Measurement”
Open an article for the complete method, evidence, limitations, and a focused reader discussion.
-
-
Microsoft’s holiday AI shopping survey measures plans, not purchases
Microsoft reports planned AI use for holiday spending. Read the survey limits and download an offer-information audit for product, price and delivery checks.
-
ChatGPT custom audiences: why uploaded rows are not matched users
ChatGPT Ads needs 25,000 matched users for inclusion targeting. Work through exclusions, overlapping lists, and bid adjustments with a count-only worksheet.
-
Two GEO defense papers test different failures in AI answers
Two September 2026 papers study GEO defenses. Compare factual distortion with source-selection manipulation, and download our evidence-review worksheet.
-
Kagi Can Hide Paywalled Search Results: What Publishers Should Measure
Kagi added a preference to remove paywalled links from results. Publishers need a matched on/off test before blaming crawling, indexing, ranking, or the filter.
-
ChatGPT Enterprise Search Can Use Cached Results: What Publishers Should Test
Restricted ChatGPT Enterprise and Edu workspaces may use cached web results or disable search. Test live, cached, and off states before making freshness claims.
-
Google’s AI Search Opt-Out Is Worldwide: Audit Property Inheritance
Google’s AI Search inclusion control is now worldwide. Audit parent and child properties before excluding content, traffic, and impressions.
-
Learn AI SEO: A Free 30-Day Evidence-First Practice Plan
Learn AI SEO through seven practical modules and a 30-day portfolio: foundations, crawler controls, claims, citations, referrals, experiments and automation.
-
AI Visibility Reports: Track Mentions, Recommendations, Citations, and Revenue
Build separate ledgers for mentions, recommendations, citations, detected referrals and business outcomes, then connect them with stable keys and explicit uncertainty.
