Q] To begin with, Agentic AI is perceived as a step towards autonomous marketing driven by AI. In such a model, where does accountability lie, and who ultimately takes responsibility for decisions made by AI systems?
It applies across industries: first comes automation, and the endgame is autonomy. But in between, we’ll always have a human-in-the-loop model. For example, in our multi-agent system, every step involves human validation; nothing is executed without approval. The system will ask, ‘Is this the right understanding?’ or ‘Is this what you want?’ allowing for clarifications/confirmations and ensuring brand guardrails. So, while the long-term vision is full autonomy, we’re currently in a phase where humans guide the process like a marketer (Gru) managing a team of AI agents (minions).
Q] As this is a relatively new segment of AI, what new skills do marketers need, and where are the current gaps?
Marketers need to go beyond creative thinking and build a deeper understanding of technology, especially AI, which evolves rapidly; every six months feels like a new generation. They must know what AI can and cannot do. This shift will also free up time for marketers to deepen customer understanding through qualitative research and data analysis. The better you know your customers, the better your AI inputs and outputs.
Additionally, marketers must identify which problems they want AI to solve, whether it’s improving lifetime value (LTV) or reducing customer acquisition cost (CAC). Ultimately, this is a reset in marketing. The fundamentals remain maximising LTV, minimising CAC, but now it’s about discovering new ways to achieve that.
Q] Which industry segments are showing the strongest interest in Agentic AI right now?
From our early discussions, BFSI and e-commerce are leading the way. We broadly classify industries into three categories: BFSI, e-commerce/D2C/e-retail, and digital migrators like Asian Paints, companies with strong offline presence that are now expanding digitally. Smaller and D2C brands are often faster adopters since they can make quick decisions and are not weighed down by legacy systems. Ultimately, adoption speed depends on two factors—data availability and leadership vision.
Q] Do you believe Agentic AI will eventually redefine or replace the roles of Creative and Performance Marketing teams?
Not replace, but redefine roles. Just as AI tools help writers and researchers work faster and better, and marketing teams will become leaner but more strategic. The human touch, empathy, creativity, understanding real-world conversations will remain irreplaceable.
In fact, the most valuable currency in the future will be human attention. AI agents will interact with AI agents, but ultimately, marketers will have to break through to real humans with emotion and storytelling.
Q] What do you see as the biggest challenges in adopting this technology?
There are two major challenges: First data silos - different teams owning fragmented data sets. Agentic AI relies on unified customer views, and without that, it can’t function effectively.
Second, marketing needs to evolve from being a cost center to a profit engine. For that, AI-driven marketing has to become a CEO-level agenda, not just a CMO initiative. Today, too much money is wasted on customer acquisition with little accountability for profitability. If marketing leaders embrace AI, technology, and data-driven decisions, they can drive both growth and profits, positioning themselves as future CEOs.
