Driving the AI shift: A thoughtful human lens

Key takeaways

• AI is evolving, in digital marketing and beyond, from narrow-task execution to generative and agentic intelligence.
• According to McKinsey’s 2024 State of AI report, 88% of marketers are using AI tools, but only 8% feel confident using them effectively, revealing a significant capability gap.
• Strategic use of AI requires strong data foundations, human oversight, and clearly defined outcomes.
• Creativity, ethics, and human perspective (still) remain irreplaceable, AI augments, not replaces.
• Generative tools like AutoGPT and large language models unlock new potential, but demand thoughtful, intentional integration.

Introduction

I often see the world through patterns,structures that are vast, dynamic, and sometimes unpredictable. It’s what draws me to quantum physics, where reality is made of unseen possibilities and potential outcomes. In that sense, AI, as I grasp it, is very similar. Quantum physics and artificial intelligence each explore unseen realms, blurring the lines between what is created and what is born. They remind us of the hidden layers behind the visible—the mystery beneath the surface of things. Together, AI and quantum physics represent a kind of substrate of potential, becoming real through perception, interaction, and intention.

Today, artificial intelligence (AI) isn’t just a futuristic concept, it’s changing how many industries operate. I’m personally following its evolution in the digital marketing and the ad technology arena. According to McKinsey, 88% of marketers now use AI tools, but only 8% feel confident using them effectively.

That confidence gap isn’t small, it’s the difference between experimenting and executing.

In my view, just as quantum physics reveals unseen layers of reality, AI is unlocking new dimensions in how we connect, optimize, and create, if we navigate it with clarity, curiosity, and technological intention. At Mars Media Group, we believe innovation must be approached thoughtfully and strategically, not reactively.

Understanding the evolution: Narrow AI, generative AI, and agentic AI

Narrow AI: Focused task execution

The first wave of AI focused on narrow, well-defined tasks. Assistants like Siri and Google Translate demonstrated how machines could perform specific functions, answering questions, translating text, with increasing accuracy.

In digital marketing, task-specific AI models have been used to support functions such as detecting fraudulent traffic patterns, recommending relevant content, and automating bidding strategies.

These systems don’t think broadly, but they are highly effective within their defined parameters. They remain the quiet backbone of many operational processes we rely on today.

Generative AI: Creating new possibilities

Generative AI has opened an entirely new frontier. Systems like ChatGPT, DALL·E, and Runway are capable of creating new content, text, images, code, audio, and beyond. Generative AI has transformed creative processes in digital marketing, enabling rapid personalization of ad assets, dynamic content generation, and faster scaling of campaigns. However, while generative AI brings speed and scale, it still requires human validation to ensure that outputs align with brand voice, strategy, and ethical standards.

Agentic AI: Autonomous decision-making

While agentic AI may feel like a relatively recent phenomenon, the core ideas, concepts and endeavours behind it, autonomous systems that plan and act independently, have existed in AI for decades. What changed in 2022 and 2023 was the emergence of more powerful building blocks: advanced large language models, integrations with external tools and APIs, and improved memory and reasoning capabilities.

These advances allowed AI systems not just to generate content, but to make decisions, take actions, and pursue defined goals with minimal human intervention.

We’ve since seen early-stage applications like AutoGPT and BabyAGI, which opened the door to broader experimentation with goal-driven AI workflows.

In digital marketing, agentic AI holds the potential to reshape campaign execution, real-time optimization, and customer engagement, but at this stage, its use still requires careful oversight to ensure outcomes remain aligned with business goals, ethical considerations, and strategic intent.

Comparing Narrow AI, Generative AI, and Agentic AI: Key differences explained

How AI bridges collaboration and improves performance

Harvard Business School found that AI significantly enhances performance: individuals using AI matched the output of teams working without it, showing AI can replicate certain benefits of human collaboration. Additionally, AI breaks down functional silos, where R&D professionals tend to suggest technical solutions and commercial professionals lean towards business-focused ideas. AI-enabled professionals produced more balanced solutions, regardless of background.

This insight highlights the power of AI in fostering better, more integrated teamwork, and why Mars Media Group focuses among other things, on implementing AI to enhance collaboration across all divisions on a day-to-day basis:

Start with strong data foundations

First-party data is essential to the success of any AI strategy. Without it, you’re relying on generic tools that can’t provide meaningful insights because they lack context. When AI doesn’t have the right data to work with, the results tend to sound like everything else out there.

When you control your data, it allows AI to work in a way that’s truly tailored to your goals. At Mars Media Group, our proprietary technology stack enables us to use insights in ways that align with the needs of our advertisers. In a world flooded with AI-generated content, much of it repetitive and indistinguishable, originality and intent matter more than ever.

Define focused, measurable use cases

AI is most effective when applied to clearly defined challenges. Whether it's optimizing bidding strategies, predicting audience behavior, or automating reporting, successful AI integration starts by solving real, measurable problems, not by applying technology for technology’s sake.

Maintain human oversight

AI excels at processing information and identifying patterns, but strategic thinking, ethical judgment, and creative insight remain uniquely human strengths. At least for now. AI-driven outputs at Mars Media Group are reviewed, validated, and integrated thoughtfully, ensuring it supports our broader strategic goals.

Don’t let AI dull our senses

As we integrate AI deeper into our workflows, in my opinion, it’s just as important to maintain the capabilities that make us uniquely human.

We’ve seen the effects of it with the dominant use of GPS. As adults, if the GPS system goes silent, we might be momentarily disoriented but eventually, we’ll probably recall how to navigate. We still remember what it means to find our way. But kids growing up today? Many have never developed that internal sense of direction. It’s not that it’s gone, it’s that it never had a chance to form, and it stays dormant.

So what happens when we overuse AI? Does humanity lose skills we once took for granted? Or do we develop new ones entirely? Will we look back and realize we’ve outsourced too much, or discover that we’ve evolved with it?

These aren’t answers I claim to have. But they’re questions I believe we should keep asking, especially now.

Foster a learning culture

The pace of AI development is accelerating. Organizations that foster a culture of curiosity, continuous learning, and open experimentation will be better equipped to evolve alongside the technology, and lead, rather than follow, in their industries.

A personal lens: Complexity, energy, and unknowns

From a personal perspective, I often draw inspiration from the spiritual world, where uncertainty isn’t a flaw, but a fundamental part of how things work and evolve. AI, too, operates in a space of probabilities, signals, and interpretation, capable of extraordinary outcomes, but never fully predictable.

This perspective helps me hold two truths at once: that AI is a powerful tool we should leverage, and that its power demands humility, reflection, and responsibility. Just like in quantum physics, the observer plays a role, shaping the outcomes through intention. In the same way, we must remain intentional in how we train, guide, and apply AI, knowing that the choices we make influence the outcomes we receive.

Lessons learned from real-world AI integration

Integrating AI into real-world operations has taught us important lessons:

• AI systems and tools make mistakes. Validation and constant testing are not optional; they are essential.
• Focusing on efficiency alone, without human judgment, can lead to missed opportunities or misguided strategies.
• The greatest value comes from blending machine-driven scale with human creativity, intuition, and leadership.
• AI accelerates execution, when the foundation is right, it can turn strategy into action faster than ever.
• Used strategically, AI doesn’t necessarily replace teams, it amplifies their impact, freeing them to focus on creative and high-value work.

Conclusion: The future is layered, dynamic, and contains multiple potentials

If there’s one thing AI has reminded me of, it’s that the future can’t be predicted, only shaped. At Mars Media Group, we’ve been building and evolving since 2005, not just with technology, but with people, creativity, and long-term relationships at the center.

We operate in a complex ecosystem, bringing together publishers, advertisers, and technology. AI now plays a role in every part of that journey, but how we choose to use it still matters most. We don’t believe in hype for hype’s sake. We believe in setting intentions, testing boldly, building responsibly, and continuing to ask the right questions as we go.

Because in a space that changes this fast, it’s not about keeping up, it’s about leading and staying ahead of the curve.