Blog

Thinking about AI visibility

Original insights on how AI talks about brands, and how to shape the conversation.

[01]

Latest

Isometric illustration of a Claude-styled terminal window with a code panel badge
ResearchFeatured

Inside Claude Code: What Leaked Source Code Reveals About AI Search Visibility

Anthropic accidentally exposed Claude Code's source through an npm source map. We went through 512,000 lines and found a domain whitelist, stricter quoting rules for most sites, mandatory citations, and a hidden blocklist. Here's what that suggests about AI visibility.

11 min readMar 31, 2026
Read article
Mack GrenfellMack Grenfell·Founder
[02]

Archive

Research·Mar 21·12 min

Why AI Picks Winners: The Research Behind LLM Brand Bias

A growing body of academic research shows that AI recommendations aren't neutral. They're systematic, measurable, and surprisingly predictable. Here's what the science says.

Research·Mar 28·10 min

Different Models, Different Favorites

ChatGPT, Gemini, and Claude don't recommend the same brands. Academic studies across 567K+ samples show systematic model-level preferences - and alignment training makes it worse.

Research·Apr 4·11 min

How to Get Recommended: What the Research Actually Says

Princeton researchers found content optimization can boost LLM visibility by 30-40%. Social proof helps. Scarcity signals hurt. Here are the tactics with evidence behind them.

Research·Apr 11·10 min

The Fragility Problem: Why AI Visibility Is Unstable

Paraphrasing a prompt can shift brand recommendations by 100%. Cold start biases lock in early favorites. Research shows AI visibility is far more fragile than search rankings ever were.

Research·Mar 23·14 min

Answer Bubbles: What Happens When AI Search Engines Disagree

A new study of 11,000 queries reveals that Search GPT and Google AI Overviews draw from fundamentally different sources, flatten uncertainty, and create invisible "information realities" that vary by platform. The implications for brand visibility are significant.

Strategy·Jan 17·10 min

ChatGPT Ads: Why the SEO/SEM Parallel Is Misleading

OpenAI is testing ads in ChatGPT. Everyone's comparing it to Google's evolution from organic to paid. They're mostly wrong, and the real opportunity is something different.

See how AI talks about your brand

Enter your domain to get a free AI visibility report in under 60 seconds.

14-day trialCancel anytime60 second setup