Articles
Research, experiments and analysis on AI visibility, generative search, SEO and how artificial intelligence systems understand the web.
Gemini Was Supposed to Hack a Fake Company. It Found Three Real Ones Instead.
The most useful lesson from the latest Gemini security incident is not that an AI suddenly decided to attack the real world. It is that an agent was given a legitimate offensive-se...
Read more →
Claude Paid a Website to Read One Page—but AI Crawlers Still Arrive Without Wallets
A website has now demonstrated something that still sounds slightly futuristic: an AI-assisted workflow encountered an HTTP payment requirement, checked its spending limit, paid fo...
Read more →
Reddit AI Search Turns Personal Experience Into Generic Advice—and Rewards the Comments That Were Already Winning
Reddit's value as an information source comes largely from something conventional search results often struggle to reproduce: people describing what actually happened to them. A ne...
Read more →
Headings and Lists Can Redistribute AI Citation Credit—but They Don’t Reliably Get a Source Into the Answer
Headings and lists may change how an AI answer engine distributes citation credit after sources have already been retrieved, but a new controlled experiment does not show that form...
Read more →
From Sardinia to the US: Tottubella.com Is Becoming Our Real-World SEO and Content Experiment
Some SEO experiments begin with a spreadsheet, a test domain and a narrowly defined hypothesis. Tottubella.com began differently: with Sardinia itself. We wanted to build a useful....
Read more →
One ChatGPT Conversation Can Now Search Across Personal and Business Accounts—Making Connected Data Part of the Answer Layer
ChatGPT can now connect more than one account from the same external service, allowing personal and business information to participate in a single conversation without forcing the...
Read more →
ChatGPT Enters Word for Free: AI Research, Rewriting and Publishing Now Happen Inside the Document
ChatGPT has moved directly into Microsoft Word, turning one of the world's most familiar document editors into another native surface for OpenAI's assistant. According to OpenAI's....
Read more →
Your AI Policy Could Become a GEO Asset—Answer Engines Need a Citable Source for How Your Brand Uses Data and AI
Companies increasingly publish privacy policies, security documentation and sustainability reports because customers, regulators and partners need authoritative answers about how t...
Read more →
Grok Doubles Down on Voice Search: Its New Transcription Model Handles Multilingual Queries at $0.10 an Hour
SpaceXAI has released Grok Voice Transcribe 2.0, a new speech-to-text model aimed at the messy audio conditions that increasingly sit between users and AI systems: multilingual voi...
Read more →
Cloudflare Can Keep robots.txt Aligned With Its Firewall—but It May Also Replace Per-Bot Control With Three Broad AI Categories
Cloudflare is trying to solve a deceptively common AI-crawler problem: a website's robots.txt says one thing while its actual firewall rules do something else. Bot Preference Sync....
Read more →
AI Content Payments Have Four Different Ranking Events: Crawled, Retrieved, Used or Judged “Significant” by the Platform
The emerging market for paying publishers when AI systems consume their work is not converging on a single unit of value. It is splitting into several. One platform can pay when a....
Read more →