Let me be blunt: I moved to paid subscriptions because I got tired of advertisers trying to tell me what I can and can’t say.

No, really. Over and over again, potential sponsors wanted “editorial alignment”—which is code for don’t say anything inconvenient.

I even walked away from a $100,000 sponsorship because I refused to play that game.

Here’s the truth: I’ve been running this at a massive loss every year because I love doing it. But love doesn’t pay for research, editing, or time. Now, readers are stepping up, saying they want to pay. They want the real story. They’re tired of sanitized trade coverage and want someone to actually pull back the curtain.

Yes, I’ll still take sponsors—but I won’t change the coverage. Ever.

If you want that kind of journalism to survive in this industry, now’s the time to support it.

The real power behind the fake AI boom in advertising isn’t the tech bros. It’s not even the agency pitch decks. It’s the money people—VC firms, PE funds, and holding companies desperate for a post-cookie growth story and high-multiple exit strategy.

And what’s hotter than AI right now? Nothing. It’s gold rush logic. AI = valuation boost. AI = acquisition bait. AI = “we’re future-proofed.”

So they AI-wash everything.

Why? Because when you say AI, you don’t have to show revenue. You just show “potential.”

And potential is all that matters when your entire plan is to inflate, pitch, and flip.

Behind the scenes, multiple adtech firms have quietly rebranded as “AI-first” without updating a single line of code. Some have bought .ai domains to look modern. Others added "AI" to job titles or product suites. None of this is real innovation. It’s packaging. It’s lipstick. It’s bait.

And it’s also possibly criminal.

Here’s why: If you’re making specific claims about AI capabilities—like fraud prevention, optimization lift, or predictive audience modeling—and those claims can’t be proven or audited, you are making material misrepresentations.

“That’s not just a marketing problem. That’s a securities fraud problem. A deceptive trade practices problem. A false advertising problem.”

And the FTC is already onto it.

In early 2024, the FTC launched Operation AI Comply in collaboration with the DOJ, CFPB, and EEOC. Their stated goal? To go after companies that are:

Sound familiar?

So far, five companies have already been hit with enforcement actions—ranging from fake AI hiring tools to bogus emotion-detection systems and unproven healthcare diagnostics. The pattern? Bold AI claims. Zero substantiation.

And here's the kicker: sources close to current investigations tell us adtech is next.

They’re already reviewing media tech platforms with vague AI promises baked into sales decks and Series B investor calls. Some firms have even been flagged for making AI performance claims in SEC filings—a legal landmine waiting to go off.

Translation? If your AI product can’t back up what you’re selling, you may be looking at actual regulatory risk, not just reputational damage.

But it gets worse: Many of these AI-enhanced products are being used in high-stakes environments—healthcare advertising, political targeting, financial services. If the AI behind that personalization isn’t real? That’s fraudulent misrepresentation with real-world consequences.

So let’s recap:

Welcome to AdTech’s Elizabeth Holmes moment.

Except instead of one company and a fake blood test, we’ve got half the industry making “AI-powered” promises they can’t explain, let alone prove. And the people funding it? They’re not innocent. They’re actively encouraging the fraud because they believe they’ll be out before the music stops.

This five-part ADOTAT+ series will pull the curtain back on the investment-fueled AI fantasy. We’ll break down:

Get ready. Because this time, it’s not just about misleading marketers.
It’s about lying to regulators.
And lying to investors.
And doing it at scale.

Stay bold. Stay skeptical. And if your AI pitch isn’t backed by actual math?
You’d better hope your lawyers are better than your model.

Welcome to the most overhyped hustle in advertising: scrape some data, match a few fields, toss in “AI” somewhere in the slide deck—and boom, valuation boost.

This is the playbook. And it’s everywhere.

If you’ve ever sat through a vendor demo that promises “predictive personalization at scale,” odds are you weren’t watching anything remotely resembling real artificial intelligence. What you got was a low-rent scraping operation dressed up in technical jargon—a glorified metadata blender claiming to be sentient.

Let’s unpack the grift.

Scraping is what it sounds like: automated bots comb through websites, APIs, PDFs, or anywhere else that data lives, hoovering up everything from names and emails to behavioral tags, product listings, and timestamps. In the adtech world, this data is then “enriched,” which usually means matched against other scraped datasets, labeled with generic attributes, and passed off as high-value intelligence.

It’s not.

This isn’t machine learning. It’s automated Excel work, just with fewer interns and more overpromises.

Let’s define our terms:

This isn’t intelligence. This is automated clerical work—done faster, not smarter.

In fact, many of the systems labeled “AI” in pitch decks are just:

No training data. No neural networks. No modeling.

Just logic trees that your dad could’ve built in Microsoft Access in 2002.

Why does this matter? Because selling this as “AI” isn’t just misleading—it might be illegal.

As part of Operation AI Comply, the FTC has said it will investigate and prosecute deceptive claims about AI capabilities, particularly when the gap between marketing and reality is wide enough to drive a Tesla through.

Translation? If your adtech company is scraping LinkedIn data, matching it with demographic buckets, and calling it “real-time predictive optimization,” you’d better be prepared to prove it. Because if not? You’re not just overhyping. You’re misrepresenting a product in a way that could expose clients and investors to regulatory risk.

This kind of fake AI persists because:

But behind the curtain, most of what’s billed as “autonomous targeting” or “AI-driven personalization” is just a glorified join between CRM data and interest-based segments. You’re not building HAL 9000. You’re mapping ZIP codes to device IDs and hoping no one audits the logic.

Let’s also talk about the ethics dumpster fire.

Scraping data en masse—especially personal data—without consent is already a privacy nightmare. When you combine that with matching metadata fields from public sources, you bypass meaningful user control and transparency.

Worse: this “AI” is often deployed in high-stakes contexts. Think financial services, healthcare ads, or political messaging. A wrong match or misclassified field doesn’t just mess up targeting—it impacts real people in real, sometimes dangerous, ways.

And no, the “but it was public!” excuse doesn’t hold. Privacy laws from GDPR to CCPA are increasingly clear: publicly available ≠ freely usable for automated decision-making.

Let’s compare this grift to actual predictive AI:

If you’re a company that matches names, ages, and ZIP codes from scraped data and labels it “artificial intelligence,” here’s your wake-up call:

This is the great metadata grift, and it’s foundational to the AI smoke-and-mirrors show that has taken over adtech. If we don’t call it out, we’ll keep feeding budgets and belief into a machine that doesn’t exist

🔒 WHAT YOU’RE MISSING IN ADOTAT+
“Everyone’s AI-First… Until You Ask for the Source Code.”

Agencies are rebranding as AI consultancies overnight—but behind the jargon is mostly ChatGPT prompts, white-labeled tools, and zero data science.

We break down:

If your agency’s AI can’t be audited, demoed, or explained—it’s not a platform. It’s a liability.

👉 Full story only in ADOTAT+
Stay Bold. Stay Curious. Know More Than You Did Yesterday.