Scalevise

Google’s 2026 Search Removal Tool Updates Expand Governance

Google updated its removal tooling and guidance in 2026 to broaden eligibility and reporting for sensitive personal data, non-consensual imagery, and AI-generated deepfakes. Industry coverage (via Search Engine Land) frames the changes as part of a three-part framework—source removal, deindexing from Google Search when eligible, and legal routes—emphasizing that deindexing affects discoverability but not host-level availability. The update introduces expanded scenarios (including government-issued IDs, passport data, and Social Security numbers) and proactive alerts, shifting organizational handling of harmful search results toward a governance workflow that coordinates privacy, legal, security, communications, and monitoring functions. The changes strengthen options for enterprises but do not replace source-level takedowns or formal legal remedies.

Google Expands Gemini Notebook with Visible Thinking Steps

Google has renamed and expanded NotebookLM into Gemini Notebook, a more integrated research workspace that connects the Gemini app, Google Search and Google Workspace. New user-facing features include expandable "thinking steps" for eligible Google AI Ultra and Pro subscribers, in-notebook code execution, and continued citation-backed outputs (Audio Overviews, Mind Maps). The rollout is phased (initial access for Google AI Ultra and Workspace AI Ultra Access, then Pro on the web) with mobile expansion planned. Google says uploaded sources remain private and are not used to train models; the company frames thinking steps as experimental and an inspection aid rather than proof of correctness.

AI Search Creates Measurement Gap Beyond Clicks

AI-driven search interfaces can cite or include brands in answers without sending users to the brand's website, creating an attribution gap for marketers. Wix Studio's AI Search Lab analyzed roughly 75,000 AI-generated answers and more than one million citations and found that listicles, articles, and product pages account for a disproportionate share of AI citations. Industry discussions and summaries (including a 25,000-URL Search Engine Land dataset) suggest content structure affects AI-search visibility. The article outlines an emerging measurement framework—metrics like AI citations, answer inclusion, Share of Model Voice, prompt coverage, and conversion influence—that complements traditional ranking/impression/click-based analytics. It recommends tracking citations and mentions separately, documenting engine/region/prompt, and aligning prompt sets with commercial priorities.