Each week we'll gather headlines and tips to keep you current with how generative AI affects PR and the world at large. If you have ideas on how to improve the newsletter, let us know!

What You Should Know 

Why Specificity Has Never Mattered More 

“Leading provider of innovative solutions.” “Strategic partner for leading brands.” “Helping organizations thrive in a dynamic environment.” These phrases and others like them show up in press releases, bylines, company websites, and executive bios every day. They sound like they’re saying something, but they’re not. Somewhere along the line, communications became more about filling space than making a point. That problem has been around a lot longer than AI.

But AI — more specifically, AI search — is making it impossible to ignore, and crucial to fix. According to Conductor, 94% of CMOs plan to increase their investment in AI search this year, and 97% said those efforts had a positive impact on their marketing funnel in 2025. Brands are spending real money to show up in AI-generated answers. The challenge is that AI search doesn’t reward the same vague, everything-to-everyone messaging that has dominated B2B communications for years. It rewards clarity and specificity.

When someone asks ChatGPT or Google’s AI Mode a question, it often carries their unique mix of context, problems to solve, and requirements to meet. The system doesn’t return a ranked list of links like traditional search engines. It synthesizes an answer, drawing on media coverage, websites, directories, and other sources — whatever it determines is most relevant. Generic content gives AI tools nothing to match against. 

A study we released last week about wealth management in AI search reinforces this. After analyzing more than 200,000 AI responses across five major platforms, we found that when users asked about advisors serving specific client types, AI systems cited firm-owned pages built around those specialties far more often than national media. The firms that described their expertise in specific terms showed up. The ones that kept it broad didn’t.

Specificity has always been what separates good communication from noise. AI is just measuring it now.

Content that says “mid-size credit unions are losing younger members because their mobile apps can’t handle real-time P2P transfers” gives an AI something concrete to cite. It also gives a human reader an immediate reason to care. Content that says “financial institutions must adapt to evolving consumer expectations” does neither. And the flood of AI-generated content online, most of it grammatically correct but completely generic, makes it even harder for vague messaging to stand out.

Every piece of content a brand produces is now potential source material for an AI answering someone’s question. If the answer to “What does this company do, and for whom?” isn’t obvious within the first few sentences, it won’t surface — in AI search or in a reader’s memory. The brands that actually have a specific story to tell will be the ones that both find compelling.

Elsewhere …

• Techie Shrinks Dog’s Tumor by Half After Using ChatGPT to Design ‘First Personalized Cancer Vaccine’
• Shopify Is Preparing for AI Shopping Agents to Change Everything, Exec Says
• OpenAI’s Latest AI Models Are Built for Speed
• AI on Deck: Assessing Impact of MLB’s New Ball-Strike System
• An Expert Guide to Using AI Tools

Tips and Tricks

💻 Get the old artifacts back

What’s happening: Claude users may have noticed that artifacts, the AI platform’s equivalent of ChatGPT’s canvas, have changed in recent months. They now default to “Markdown” files, a markup language for plain text documents that AI tools can easily read. 

What’s the difference?: There were two particularly useful elements of the “old” artifacts. First was a dropdown menu that showed previous versions of the artifact if you were iterating. Second was the ability to select text in the artifact, which would produce two clickable buttons: “Improve,” which brought up a text box where you could describe what you want to see changed, and “Explain,” which would automatically prompt Claude to provide more detail on why it wrote the selected text.

They’re not gone forever: You can actually get these more interactive artifacts back by changing your settings. If you go to the “Capabilities” tab and switch “Code execution and file creation” to off, the old-style artifacts will again become the default.

Yes, but: You may want to toggle that setting back and forth, because having the capability turned on is what allows Claude to create documents like Microsoft Word or PowerPoint files. It all depends on your preferred method for generating content drafts.

Quote of the Week

“I am not sure ‘Forward Deployed AI Engineers’ are going to deliver on what a lot of companies are hoping for. They are useful, yes, but AI applications are far less of a technical issue, and much more about rethinking the deep expertise & structure of your organization around AI.”

— Ethan Mollick, Professor at the Wharton School of the University of Pennsylvania, in a social post on AI adoption

How Was This Newsletter?