How AI is revolutionizing AdTech and MarTech: reality vs. hype – part 2

In the last post, I got carried away imagining AI as the magical genie of AdTech and MarTech, whisking us off to sunny beaches with endless drinks.🍹🌴
Today… welcome to the cold, hard reality.

Beacuse not every AI story is a slam dunk. For every real win, there’s a handful of overblown promises. Here’s where the hype creeps in.

The hype: myths and overblown claims

Let’s check spaces where AI still outplaces reality:

1. AI can prepare fully autonomous campaigns (Agentic AI)

The idea that AI can fully run and optimize multi-channel campaigns without human intervention is still largely aspirational. AI isn’t plug-and-play. Many vendors pitch AI as a one-size-fits-all solution. In reality, AI needs good data, thoughtful setup, and ongoing oversight. If you expect to “set it and forget it,” you’ll be disappointed. Current AI excels at automation, not autonomous strategic decision-making.

Why?

AI is only as good as the data it’s trained on. When the data is incomplete, biased, outdated, or skewed, the AI can make flawed or unethical decisions. 

You won’t prepare financial statements without a set of numbers.

You won’t sell clothes without a price.

You won’t get pregnant without… ok, let’s skip it.

Think about GIGO (Garbage In, Garbage Out) when poor training data leads to bad predictions and optimization. Another thing is bias reinforcement: AI may inadvertently perpetuate demographic, cultural, or behavioral biases. In 2019, a financial services company was found to be unintentionally offering less favorable credit terms to women because the AI model was trained on historical data that reflected societal biases.

2. AI can be creative

AI can optimize impressions and clicks but still has issues with contextual nuance (irony, sarcasm, emerging social controversies), cultural sensitivity and real-time brand reputation threats (yeah, everything is about brand safety). A similar case is with creativity: AI is still pattern-matching, not thinking. Can AI invent an advertising slogan like “Think Different” and match appropriately selected images (in this case black and white portraits of historical figures like Albert Einstein, Mahatma Gandhi, and John Lennon)? I don’t think so. Therefore, don’t fire your copywriter just yet.

AI doesn’t create from nothing. It synthesizes existing patterns from training data. It cannot invent entirely new cultural ideas, break creative conventions or develop disruptive, out-of-category campaigns.

Especially enterprise-level creative decisions carry significant financial, reputational, and legal risks. AI cannot take ownership for controversial decisions, understand long-term brand equity and assess non-quantifiable business impacts.

When Pepsi released its controversial Kendall Jenner ad in 2017, the issue wasn’t poor optimization – it was a strategic misstep in cultural sensitivity. AI wouldn’t have prevented it, nor could it have navigated the fallout.

AI-generated content tends to repeat winning formulas rather than invent new ones.
This can quickly lead to creative stagnation and audience fatigue.

AI still needs:

• Strong prompts -> human-created
• Quality training data -> human-approved
• Expert supervision and refinement -> human-guided

Without these, AI-generated creative output can feel generic, misaligned, or off-brand. Moreover, AI can inadvertently generate offensive, misleading, or legally risky content if not carefully monitored.

3. AI can be compliance with Universal Data Privacy

AI does not inherently understand data privacy regulations (CCPA, CPRA, GDPR, HIPAA etc.) and industry-specific legal constraints (financial disclosures, healthcare advertising rules). AI isn’t a smart, fast and privacy-friendly ninja with a GDPR manual.  Facebook’s ad targeting algorithms had to be reined in because they enabled discriminatory housing and employment targeting, inadvertently violating U.S. law.

Legal teams and marketing leaders must supervise AI models to ensure campaigns comply with local, national, and industry-specific laws.

AI isn’t a silver bullet for data privacy. Many enterprises mistakenly think AI can “fix” compliance. In reality, AI systems must be designed with privacy-first architectures – especially when navigating state-specific laws.

And you need to be aware that even the best AI models decay over time due to shifts in consumer behavior, new competitors entering the market or seasonality changes.

Common myths about AI

Doing my research I have come across some AI bullshits that do not only apply to the Adtech and Martech industry. Let’s deal with common myths: 

First, the myth is that AI guarantees instant marketing success. Plug in an AI tool and watch the leads roll in? Not quite. AI can make your campaigns smarter, but it won’t magically fix bad strategy or poor creativity. Results still take work and time. 

Another myth is that AI will replace marketers. AI is a tool, not a replacement for human creativity, strategy, or judgment. The best results come when marketers use AI to enhance, not replace, their own expertise. 

Some believe all AI is created equal. That’s not true. Some tools are far more advanced than others. Don’t fall for marketing fluff. Ask tough questions about how the AI works, what data it uses, and what results you can expect.

Finally, there’s the fear that AI will make your job obsolete. AI is here to help you work smarter, not to take your job. The best marketers are learning how to use AI as a partner, not a threat. 

AI is powerful, but it’s not a silver bullet. The smartest marketers know how to separate the real from the ridiculous. The real competitive edge lies in custom AI-powered AdTech and MarTech solutions tailored to your specific needs, data environment, and compliance landscape. Off-the-shelf platforms rarely fit enterprise complexity. Custom solutions are safer, smarter, and more scalable.