In partnership with
You Were Always the Product
Remember when ChatGPT felt like the one clean place on the internet?
That's ending this month.
OpenAI announced yesterday that visual ads are coming to ChatGPT image generation.
You ask for a picture.
While it renders, a labeled display ad appears alongside it.
Products, services, "inspiration," according to OpenAI.
It starts later this month, US only, with a test group of advertisers.
And it's aimed at the free and low-cost tiers, which is where most of ChatGPT's 1.2 billion weekly users live.
Now, OpenAI says all the right things.
The ads are clearly labeled.
They stay separate from the image you're creating.
They don't influence ChatGPT's answers.
Fine. I'll take them at their word on all three.
Here's the part that deserves a longer look.
This is a reversal.
Sam Altman once called ads inside an AI product "uniquely unsettling."
He described them as the thing you do last, when nothing else works.
As recently as December, the head of ChatGPT was denying any live ad tests.
Ten months later, OpenAI is building a complete advertising business.
Measurement partners. Brand-safety vendors. Geo-targeted experiments. Attribution tools.
Those aren't the moves of a company dipping a toe in.
That's an ad platform assembling itself.
And here's the twist that should make you pay attention.
OpenAI is now both the advertising platform AND a competitor on it.
Reports say it told partners it would no longer accept ads for rival image and voice generation products.
The same week it launched its own new image model and its own live voice product.
So the categories where it sells ad space are the categories where it's quietly narrowing the paths for anyone who competes with it.
Read that as what it is.
The free tier is being paid for by advertisers.
Which means the free tier's real customer is the advertiser, not you.
That's not a scandal. It's how every free product in history has worked.
But it changes how you should think about the tool.
If you build workflows on the free plan, you're building on a surface that's being optimised for someone else's goals.
If you pay, you're buying the version where you're still the customer.
And if you run your own models, you're the only one in the room.
Meta's Muse being free to everyone is the pressure that forced this.
The price war I've been tracking all summer didn't just make models cheaper.
It made "free" expensive in a different currency.
Decide which currency you'd rather pay in. 🫡
Did You Know?
The first banner ad on the internet ran in 1994, for AT&T, on HotWired...
Its click-through rate was 44%.
Today the average banner gets clicked about 0.05% of the time.
Thirty-two years of ads taught the whole internet to look away.
OpenAI's betting you haven't learned yet. 👁️
Blu Dot surpasses 2,000% ROAS with self-serve CTV ads
Home furniture brand Blu Dot blew up on CTV with help from Roku Ads Manager. Here’s how:
After a test campaign reached 211,000 households and achieved 1,010% ROAS, the brand went all in to promote its annual sales event. It removed age and income constraints to expand reach and shifted budget to custom audiences and retargeting, where intent was strongest.
The results speak for themselves. As Blu Dot increased their investment by 10x, ROAS jumped to 2,308% and more page-view conversions surpassed 50,000.
“For CTV campaigns, Roku has been a top performer,” said Claire Folkestad, Paid Media Strategist, Blu Dot. “Comping to our other platforms, we have seen really strong ROAS… and highly efficient CPMs, lower than any other CTV partner we've worked with.”
Using Roku Ads Manager, the campaign moved from a pilot to a permanent performance engine for the brand.
🗞️ The Tip AI News 🗞️
Nvidia wants to finance AI like airplanes.
Follow the money for a minute, because it explains why your tools are changing so fast.
Nvidia is working on a plan to turn its chips into collateral.
The idea is that AI developers could borrow money to access Nvidia's GPUs and computing power...
With the chips themselves backing the loan, the way aircraft back airline financing.
The target is up to $500 billion in AI infrastructure, funded this way.
Jensen Huang wants compute to become a real financial asset.
Something investors can own a slice of, like a fleet of jets.
Now the other half of the picture.
Goldman Sachs says AI-related companies with weak credit ratings have already issued $88 billion in debt this year.
High-yield bonds. Leveraged loans. The risky end of the market.
And investors are starting to ask the obvious question.
Where's the revenue that pays all this back?
Here's why this matters to you, even if you'll never see a bond.
When an industry is financed with this much borrowed money, every company in it has to show returns fast.
Which is exactly why you're watching the same pattern everywhere.
Ads in ChatGPT. Pro plans halved. Flagships behind credit meters. "Unlimited" quietly becoming "metered."
None of that is random.
It's what a borrowed-money boom looks like when the bills start arriving.
The models will keep getting cheaper per token, because competition forces it.
The free rides will keep getting shorter, because the lenders demand it.
Plan your stack for both. 🫡
Over to You...
Go check which tier you're actually on, and ask who it's really serving...
Then go decide which currency you want to pay in. 🔥
Founder, AI Persona Method | TheTip.ai
Get paid to solve problems like this → AI Certified Consultant
PS. Mizuho, one of Japan's biggest banks, is rolling Claude Cowork out to 30,000 employees for non-routine tasks... While the free tiers get ads, the enterprises get agents. Two different products, same company. 👀
» Join the AI Money Group «
💰 AI Money Blueprint: Your First $1K with AI - Learn the 7 proven ways to make money with AI right now
🚀 Zero to Product Masterclass - Watch us build a sellable AI product LIVE, then do it yourself
📞 Monthly Group Calls - Live training, Q&A, and strategy sessions with Jeff
Sent to: {{ email }}
Jeff J Hunter, 3220 W Monte Vista Ave #105, Turlock,
CA 95380, United States
Don't want future emails?
