🤓 Why ChatGPT Recommended My Newsletter (and What It Means for You)
PLUS: An AI artist lands multi-million-dollar deal; Getty loses big copyright lawsuit—and more
👻 Halloween may be over, but AI’s still out here pulling off the costume—and exposing the gap between marketing and reality.
If you work in entertainment, media, marketing or PR and missed last week’s edition, you’ll want to read it.
---
Monday morning, 3:37 AM.
I got a notification that someone subscribed to this newsletter.
Source: chatgpt.com
The email address was something like “XXXagent@gmail.com” (details obscured for privacy, but you get the idea).
That’s likely not a person. It’s an agent. (If you’re reading this, hi)
Someone (let’s call him Adam) created an AI agent to find credible AI sources for insights.
And that agent chose my newsletter.
How an AI Agent Chooses
Here’s what probably happened:
Adam gave ChatGPT a goal, maybe something like:
“Find me thoughtful analysis on how AI is reshaping media and entertainment strategy—from someone who actually understands both worlds and brings original perspectives.”
Or maybe it was a request more aligned with a specific post, like how to get more creative responses from AI.
ChatGPT went hunting.
It scanned everything it could access.
Evaluated credibility signals I can only guess at—clarity, specificity, depth, consistency.
Whether the author actually knows what they’re talking about or just regurgitating content from major outlets.
It surfaced my newsletter—likely alongside other options—as a strong match for Adam’s personalized criteria.
It read through my archives.
Compared it with others.
Made a judgment call about quality, relevance, and fit.
Then subscribed.
What Got Me Recommended (And What it Means for You)
My newsletter is growing slowly but currently only has 553 subscribers.
(Still feels surreal. I sometimes imagine that many smart people in a room listening to me talk about AI every week. That’s still a big room.)
Meanwhile, there are hundreds (maybe thousands) of people who write about AI.
Many with far bigger audiences.
Many backed by major publications like TechCrunch, The Verge, or The Wall Street Journal with massive reach and built-in credibility.
So why did ChatGPT surface this one?
I’ve spent the past year thinking about this question: what makes brands discoverable and recommendable in an AI-mediated world.
Now I’m living it.
Here are the things that worked in my favor.
1️⃣ Clarity + specificity: Giving AI a reason to choose you
There are hundreds of AI newsletters. There are hundreds of entertainment industry analysts.
Very few people are writing at the intersection of AI strategy, media industry dynamics, shifting audience behavior, and brand implications—and even fewer actually understand both the tech and the industry it’s disrupting.
That specificity made me easier for AI to identify, and more likely to be recommended.
The same principle applies to brands.
If you want to be discoverable by AI agents, you have to make it obvious what you do, who you’re for, and what sets you apart.
The information about your products, services, or content needs to be accurate, explicit, and complete in a way AI can actually understand, so it can confidently match you to a query.
Not in the usual vague language like “industry leader” or “driven by innovation”, but in real, concrete terms that set what you offer apart.
2️⃣ Accessibility + repetition: Being visible where AI looks for its answers
I’ve been writing about AI for over a year on Substack. This is the 61st edition.
Since my content is not gated behind a paywall, it’s accessible to AI search engines like ChatGPT, Gemini and Perplexity.
I’ve also (mostly) optimized my website and newsletter editions for AI search and posted consistently on LinkedIn for over two years.
The repetition across platforms builds credibility.
When AI sees the same positioning and expertise showing up consistently across multiple accessible sources, it can trust that information.
For brands, this challenge is exponentially more complex.
Your story and messaging lives across teams, channels, spokespeople and audience touchpoints.
And agents don’t just check your website. They cross-check everything—YouTube, Reddit, product reviews, Wikipedia, news coverage, social media footprints.
They’re not looking for your official narrative.
They’re looking for alignment—evidence across trusted sources that confirms you’re the best match for the request.
That means:
The information on your own site and channels needs to be properly structured and accessible so AI agents can crawl and understand it.
Your core message (what you offer, how it works, who it’s for) needs to be repeated across the broader ecosystem of the internet—everywhere AI looks for answers about you.
The Agent Economy Is Here
Adam’s agent isn’t a one-off.
This is where we’re headed.
People are already building AI agents to:
Find products that match their needs and tastes
Surface insights from sources they’ve never heard of
Recommend restaurants, books, movies, TV shows and podcasts based on preferences they can barely articulate themselves
Soon, agents will be booking travel, making purchase decisions, negotiating deals, managing calendars, and curating the content we consume.
This is both liberating and terrifying.
Liberating, because smaller players—people like me, brands with a fraction of the resources—can compete on relevance instead of scale.
My 553 subscribers matter just as much as someone else’s 500,000—if I’m the better answer to the question being asked.
Terrifying, because if you don’t have clear positioning, if your information isn’t accessible, if your brand isn’t credible across the places AI actually checks, you’ll be invisible.
Which means your job as a brand isn’t to just show up in search results anymore.
It’s being agent-worthy—the kind of brand AI would confidently recommend when making decisions on someone else’s behalf.
This week, OpenAI CEO Sam Altman put it this way:
“You ask ChatGPT for the best hotel, not Google or something else. If ChatGPT were accepting payment to put a worse hotel above a better hotel, that’s probably catastrophic for your relationship with ChatGPT.”
AI agents are looking for the best answer to very specific questions.
And if that’s you, they’ll find you.
Even at 3:37 in the morning.
—
P.S. This newsletter isn’t paywalled (yet) because I want it to be easy to access, share, and recommend. But to those of you supporting it anyway.… I see you. And I’m genuinely grateful. 🙏 You’re helping me grow it into something bigger and even more valuable.
P.P.S. Skipping next week’s edition (I keep saying this, I know) to make room for client work and test some new capabilities. But if something big drops and I can’t help myself, you’ll probably hear from me. Otherwise, see you in two weeks.
📈 Work with Me
AI Advising and Consulting: Strategic guidance to build sustainable AI strategies and adapt to shifting audience behavior
AI Training: Practical training that helps teams use AI more strategically, effectively, and safely—so it becomes a trusted thought partner, not just a task assistant
What You Need to Know About AI This Week ⚡
Clickable links appear underlined in emails and in orange in the Substack app.
🎶 An AI singer just broke into Billboard’s R&B charts.
Last week I covered why AI music is proving to be a legit business.
This week, we have another proof point.
“How Was I Supposed to Know” by AI singer Xania Monet debuted at #30 on Billboard’s Adult R&B Airplay chart, marking the first time an AI-powered act has broken into U.S. radio airplay.
It’s also pulled in over 5 million Spotify streams,
That success reportedly sparked a bidding war among labels. She’s now signed a multimillion-dollar deal with Hallwood Media.
Xania Monet is the stage name for Telisha Jones.
You’ll see a lot of headlines describing her as a poet who writes the lyrics and feeds them into AI music generator Suno—a story that’s been widely repeated because it’s the project’s official narrative.
But after last week’s post about how AI can surface the truth behind the spin, I wanted to test that in real time. So I asked ChatGPT to run a deep dive on Jones’s background.
What is documented: she owned a print design business and sold courses teaching others how to create “AI twin influencers” and faceless brands prior to 2025.
There’s no public record of any poetry under her name before this project—though of course, that doesn’t rule out private work.
Either way, the packaging is working. Her management team handles distribution and pushes out 3–5 AI-generated performance Reels a week to drive reach and engagement.
Incidentally, fans mostly don’t seem to know—or care—that she’s AI.
And they really like the lyrics.
⚖️ AI companies get a big win as Getty Images loses Stability lawsuit in the UK.
Getty Images—the stock-photo giant behind much of the world’s licensed imagery—lost most of its UK lawsuit against Stability AI, the company behind the popular image-generation tool Stable Diffusion.
The judge found that because the model “does not store or reproduce any Copyright Works (and has never done so),” it doesn’t violate copyright under UK law.
For now, AI firms get breathing room, while creators are left pushing for clearer rules.
📱 OpenAI’s Sora is now available on Android. You can download it here.
📉 People Inc. just signed its second AI licensing deal—this time with Microsoft—after an earlier deal with OpenAI.
The move comes as Google’s AI search has cut its traffic by more than half.
🍎 Apple is nearing a deal with Google to pay the tech giant roughly $1 billion a year for a custom version of its Gemini AI model to power its overhaul of Siri.
🤝 Snap is partnering with Perplexity to integrate its AI into Snapchat starting in 2026, with Perplexity paying $400M to reach the platform’s nearly 1B monthly users.
In case you missed last week’s edition, you can find it 👇:
That's all for this week. See you either in 2 weeks or next Friday (if my schedule allows.)
Thoughts, feedback and questions are always welcome and much appreciated. Shoot me a note at avi@joinsavvyavi.com.
Stay curious,
Avi
💙💙💙 P.S. A huge thank you to my paid subscribers and those of you who share this newsletter with curious friends and coworkers. It takes me about 20+ hours each week to research, curate, simplify the complex, and write this newsletter. So, your support means the world to me, as it helps me make this process sustainable (almost 😄).