Agentic AI Has Entered the Chat: The Transformation Marketers Can’t Ignore

Watch the webinar recording now on YouTube

AI is on the verge of revolutionizing every industry, sparking a fierce competition to lead the way. For marketers, before you finalize your AI transformation strategy, there’s one crucial innovation you must understand: agentic AI. 

While everyone’s talking about productivity gains and efficiency boosts from current AI tools, the real transformation lies in systems that don’t just respond to prompts but take independent actions on your behalf. The agentic AI evolution is going to change everything about how you plan, execute, and scale AI in your marketing organization.

This was the topic of discussion in a recent Skai webinar titled Agentic AI Has Entered the Chat: The Transformation Marketers Can’t Ignore, where Chief Product Officer Guy Cohen explored the challenges of today’s AI transformation and the game-changing evolution toward truly autonomous AI agents.

Watch the webinar recording now on YouTube

GenAI is moving faster than any previous technology transformation

The urgency is real. Imagine two companies in the same category, with the same size and revenue. But one gets to their GenAI transformation first. Which one would you rather be?

But what makes GenAI fundamentally different from previous technology shifts is the unprecedented pace of adoption. “Everyone is talking about GenAI these days,” Cohen explains. “We haven’t seen in history a change like this happening so fast.”

The numbers tell the story: while smartphones took five years to reach 100 million users, ChatGPT achieved this milestone in just two months. Today, approximately 1.5 billion users are using GenAI weekly.

ChatGPT reached 100 million users in just 2 months, compared to 5 years for smartphones. This isn’t just fast adoption, it’s unprecedented acceleration that makes traditional transformation timelines obsolete.

“We have seen in the last decades a few significant transformations,” Cohen reflects, comparing GenAI to previous shifts. “It took digital transformation twenty years, it took the cloud twelve to fifteen years, it took mobile seven years to really be embedded in organizations.” But GenAI follows different rules. “What we’re experiencing now is definitely different, and I think the main thing that is different is the pace,” Cohen emphasizes. “Last week, we heard three giant announcements of new technologies with GPT-5, Gemini, and agents.”

This speed creates both opportunity and risk. Organizations that moved slowly during their digital transformation often fell permanently behind. Cohen explains that this is where there’s opportunity to outpace your rivals and gain a market advantage as a fully AI-led company. “We definitely see that not all organizations are treating it the same,” he observes. “Adoption is happening at different levels within organizations, within departments, and even within individuals. Today’s companies must move faster with their AI transformations.”

From Skai’s 2025 ShopAble event

Strategic planning separates winners from stragglers

Despite the pressure to move quickly, Cohen has identified strategic planning as the critical success factor. “We have been working on this type of transformation with more than a hundred companies around the world,” he shares. “What we have seen is that if there is one thing that has a high correlation to success, it is around learning.”

Organizations that implemented GenAI tools ad hoc consistently failed to achieve meaningful results. “It doesn’t really matter what the plan is,” Cohen emphasizes. “You have to have a plan. This plan should create a strategy. This strategy should have clear goals and milestones, accountable KPIs, and leadership investment and buy-in.”

Success requires both executive leadership setting direction and grassroots champions driving adoption. One without the other consistently fails.

The most successful implementations use a multi-directional approach combining top-down leadership with grassroots adoption. “First is top down, when the company’s leadership understands that GenAI is a technology that the organization should use. But it’s not enough. You also need to identify the change agents, these users who will jump in and use it,” Cohen explains.

However, two major factors complicate even well-planned transformations.

The human factor remains the biggest bottleneck. “The number one challenge is humans,” Cohen states. “We as humans are not comfortable with significant transformation. GenAI requires all of us to change our habits; to change things that we have been doing for years.”

The statistics reveal the scope: 87% of organizations cite a lack of understanding of GenAI as a major barrier, while 56% struggle with a lack of necessary skills and training. “Since everything is moving so fast, there is always a feeling that no one really understands what GenAI is and what they can get from it,” Cohen observes.

New risks require entirely new management approaches. GenAI introduces risk categories that didn’t exist in previous transformations. “GenAI brings with it new areas that we haven’t experienced in the past, like hallucination,” Cohen notes. “Today, it’s always a question of whether we can or cannot trust the answers that we receive.”

These risks extend beyond accuracy to include bias in outputs, data privacy concerns, cybersecurity threats, and challenges related to regulatory compliance.

The next evolution: Understanding agentic AI

While mastering current GenAI capabilities is important, long-term success requires understanding where technology is heading: agentic AI, which involves systems that not only respond to prompts but also take independent actions.

“To understand this innovation, you must first consider that the root word of agentic AI is agent. An agent is a person who acts on behalf of another person,” Cohen explains, drawing parallels to travel or real estate agents. The travel industry provides a perfect analogy: “We started with human travel agents, then companies like Expedia came and democratized travel booking. The next generation is definitely around the agentic era, companies like WonderPlan that are not only providing insights about your itinerary, but they’re actually learning about you as a consumer and booking flights, hotels, and experiences for you.”

Cohen identifies two categories: Horizontal Agents that provide broad capabilities across the organization (like ChatGPT or Gemini), and Vertical Agents that specialize in specific domains with deep industry knowledge.

Why vertical agents represent the real breakthrough

Cohen makes it clear that the future of marketing lies within agentic AI, “Marketers will continue to use horizontal agents to write emails, find facts, etc., but the real value for marketers will come in the form of agentic AI solutions that can deliver on the true value of positively impacting efficiency, consumer relationships, and, most importantly, the bottom line.”

For vertical agents to deliver transformative value, organizations must establish four fundamental capabilities:

Solid data foundation. “Agents need to work on clean, connected, contextualized data that is as broad as possible since the agent is as good as the data that it has access to,” Cohen explains. This encompasses real-time integration, standardization, and accessibility across all relevant business systems.

Personalization and memory. “An agent for a marketing department is different from an agent for the accounting department. This agent needs to have memory, what the agent needs to remember and forget, for how long, at what level,” Cohen describes. Effective agents must understand organizational context and adapt to specific business constraints.

Integration across systems and agents. “This agent also needs to connect to other agents, there are protocols like MCP that allow agents to talk with other agents and expand their capabilities,” Cohen notes. No business function operates in isolation.

Controllable autonomy. “We are looking at agents that are doing some tasks autonomously. What’s really important is that the user is controlling the autonomous level of activity,” Cohen emphasizes. Organizations need the ability to gradually expand agent authority while maintaining oversight.

Three phases toward true agentic AI

The transition to truly agentic AI unfolds through three distinct phases as both technology capabilities and organizational readiness mature.

The journey from assisted to autonomous agents isn’t just a technology evolution, it’s a fundamental shift in how marketing work gets done. Most organizations are still in Phase 1.

Phase 1: Assisted Agents function as intelligent assistants. “Conversational interface, linear interaction, but humans are still driving,” Cohen explains. Marketers can ask complex questions and receive sophisticated analysis, but must specify exactly what they want. “Already here we are seeing enormous efficiency gains of around 40%,” Cohen notes.

Phase 2: Personalized Agents introduce context awareness and learning capabilities. “This agent actually knows you, knows the organization, and understands your business goals and constraints,” Cohen describes. These agents maintain memory across conversations and adapt recommendations based on the preferences they have learned. “The level of answers you will get is significantly better because it’s tailored for you,” Cohen emphasizes.

Phase 3: Autonomous Agents represent the full realization of agentic AI. “This is when we will be able to rely on agents to carry routine tasks for us,” Cohen envisions. “Think about your monthly report. From now on, you can ask that agent to send it to all participants on the first day of every month without your involvement.” These agents handle entire workflows, escalating to humans only when needed.

This evolution means marketers will transition from campaign operators to strategic directors, focusing on defining success metrics and creative direction rather than managing execution mechanics. Teams will need fewer platform specialists and more experts in customer psychology and competitive analysis.

The future of marketing work: agentic AI handling routine execution while humans focus on strategy, creativity, and high-level decision making.

Watch the webinar recording now on YouTube

Conclusion: The transformation starts now

As Raheel Khan, SVP of Enterprise Marketing and Head of AI at Estée Lauder, recently observed: “In the fast-moving world of GenAI, six months from now is actually today. If you’re not adopting now, you’re not just standing still, you’re falling behind.”

Cohen’s advice cuts straight to the point: “Don’t hesitate to jump, don’t get left behind, but plan in advance. Don’t do it without a plan. Plan and strategize your GenAI transformation, and I am sure that you’ll see a lot of great value.”

Cohen concluded the webinar by showcasing Skai’s agentic AI, Celeste, the only GenAI agent designed specifically for commerce media. The demonstration of this agentic AI showed how these concepts are already transitioning from theory to practice, with agents that can analyze campaign performance across hundreds of retailers, identify optimization opportunities, and provide personalized recommendations tailored to your specific business goals and constraints.

The companies that build strategic AI capabilities now, while understanding the trajectory toward agentic systems, will define the competitive landscape for years to come. Skai is the AI-driven commerce media platform for performance advertising, and with Celeste AI, marketers can experience the future of agentic marketing today while building toward even more autonomous capabilities.

Ready to explore how agentic AI can transform your marketing organization? Schedule a quick demo.

---