The AI Promo Brief: Issue #23

Getting Chosen by AI Is Only Half the Battle

Go ask ChatGPT or Gemini right now and search for your own company. Did it show up? Congrats, you've won exactly half the battle. The other half is trickier. Search Engine Land's breakdown of a 30-million-source AI-citation study found the single most-cited source AI answers pull from is Reddit (with YouTube and Wikipedia not far behind). Not your homepage, not your slick product page, but community threads where real people argue about who's actually reliable. And it gets harder: a new "Credibility Paradox" analysis from Burson found that being mentioned in an AI answer doesn't mean the mention reads as believable. The smart play: stop polishing only your own copy. The conversation about your brand in communities — and whether you are trustworthy — is now part of your search footprint. Go listen to it.

How Far Behind Is Your Team, Really?

Be honest: do you actually know whether your team is ahead of the competition when it comes to AI, or are you just hoping? The Marketing AI Institute's 2026 State of AI for Business report gives you a yardstick: listen to how 2,100+ marketing pros, 84% of them B2B, think AI is reshaping the market right now. It's the closest thing to a peer benchmark you'll get, and the companion breakdown digs into what B2B marketers themselves are saying about where they stand. The smart play: read it as a scorecard. If your team's barely past ChatGPT-for-subject-lines while the median B2B team is restructuring workflows, that's your gap.

Adobe's Brand Visibility Tool Closes the Loop

Soon your brand's AI reputation will police itself — at least if you're paying Adobe for the privilege. Brand Visibility tracks how ChatGPT, Perplexity, and Google's AI Mode describe you, then loops that straight into Adobe's stack: Firefly and GenStudio draft the fix, Experience Manager pushes it live to the edge, the models pull the new story without a republish. A self-correcting brand machine. The honest footnote is that the loop only closes if you already live inside Adobe's walls —  and the same company grading your AI reputation is the one selling the fix.

Google Just Made the Language Barrier Optional

Google just shipped Gemini 3.5 Live Translate, real-time speech-to-speech across 70+ languages, live now in Google Translate and rolling into Google Meet. The thing that matters: it's natural-sounding and continuous. It's not the choppy, wait-for-the-whole-sentence cadence that made the old tools useless on a real call. For overseas sourcing or an international client, that means fewer dropped nuances on pricing, lead times, and decoration specs. It won't replace your trusted importer on a high-stakes project — but for the routine back-and-forth, it quietly removes a friction point you'd stopped noticing. One to watch.

The AI Automation Driving 20% Growth

How much of your sellers' week vanishes into research that isn't selling, like pulling order history, prepping for the call, chasing down what a client bought last spring? JPMorgan just put a number on it: AI tools performing overnight client and market reviews have lifted private banking gross sales 20%. And that's before the bigger move: the bank is now rolling out longer-running agents that work for hours without a human babysitting them. Read that carefully: the agents aren't closing anything. They're doing the busywork so a banker walks into every meeting already briefed, with more hours to spend on the actual relationship. That's the lesson for any promo pro sizing up agent tools. The win isn't an agent that pitches your client — it's an agent that owns the prep, the data pulls, the digging, so that your reps show up sharper and sell more. Let AI replace your time, not your handshake.

Meet the AI Description Re-write Tool

Ever pasted a supplier description into a presentation and watched the jargon do the talking? commonsku's new AI Description Re-write Tool turns that spec-heavy copy into language your client reads in one go, serving up multiple client-ready variations from existing product info right inside the form. It's insanely easy and fast:  Choose one, tweak it, close, done. It's free in open beta for Advanced and Enterprise plans and your clients get pitches that sound like they were written for the human merch experience, not a list of material specs!

OpenAI's Code Red Is Your Cue to Stop Overcommitting

How many AI subscriptions is your team quietly stacking right now? One for chat, one for the coders, one bolted into the browser? OpenAI declared an internal code red back in March and said it would merge ChatGPT, Codex, and its Atlas browser into a single desktop superapp. And as of early-June reporting, that merger is landing within weeks, with Greg Brockman running product and Fidji Simo driving the commercial push. Translation: the company that fragmented your AI stack is now actually un-fragmenting it, not just talking about it. The market is consolidating into fewer apps.

⚡ Blink and it was gone: Anthropic launched its most powerful model yet, Fable 5, then a first-ever U.S. export order yanked it offline days later. 

🤖 SpaceX is acquiring Cursor for $60B in all-stock, days after its own IPO — eye-watering, but it's a flare for where the money thinks agentic coding is headed.

🧠 The AI talent wars are wild: Anthropic grabbed OpenAI co-founder Andrej Karpathy last month, then a Nobel-winning DeepMind scientist this week. Claude's lab is stacking the bench.

⚖️ Trump advisers are weighing a government equity stake in major AI companies, with Treasury and Commerce pitching different structures. Nothing's decided, but the rules your AI tools live under are being drawn right now.

💰 Ramp raised $750M at a $44B valuation and shipped a corporate card built for AI agents to spend on a company's behalf. Agentic procurement just went mainstream.

🪑 The AI Round Table is where you can share and hear how your peers are using AI, pull up a chair with host Bobby Lehew when you're ready to join us!