All in on AI: “The Ultimate Amplifier”

In January, 2026 we launched a new leaders series called “All in on AI”. In part one, I wrote about the Future of Marketing in APAC and how AI is reshaping our industry.

To kick off our industry leader conversations, we turned to a theme that keeps coming up across markets: AI as the ultimate amplifier.

We sat down with one of the region’s most respected AI voices, Kellyn Coetzee, Sydney Head of Digital at Zenith Australia and Co-Chair of the IAB Australia AI Working Group. Kellyn shared her perspective on how AI can empower every marketer to achieve more, along with the key shifts she believes will define the year ahead:

Responses attributed to Kellyn Coetzee, Sydney Head of Digital at Zenith Australia and Co-Chair of the IAB Australia AI Working Group

How are you and your agency [Zenith] using AI right now to strengthen existing capabilities or marketing investments and what impact are you seeing so far?

We are focusing our efforts on two main areas: AI for the individual and AI at scale. The former aims to democratise tools, skills, and capabilities across the agency, making them accessible for everyone's unique roles and needs. The latter involves operationalising AI to manage the volume and velocity of media buying.

Lately, I've observed many instances of ‘automation’ being labelled as AI, with large companies promoting grand AI tools that have little practical application at the ground level. While I can't delve too far into the specifics of what we’re doing, I believe what sets us apart is our integrated approach. This makes me incredibly excited about the future.

Where does AI genuinely improve the way you work, whether in briefing, ideation, or creative versioning? And where do humans still need to lead?

We are highly focused on AI ‘augmentation’ – utilising AI to accelerate specific segments of the media cycle. By addressing bottlenecks and challenges within our workflow, we can meet the needs of our clients and vendors through optimising our performance.

AI helps us bridge gaps, enhance our products, and allow our teams to shift their focus from "How am I going to get this all done?" to "How can I make this better?" This transition provides more time for strategic thinking and innovation.

As AI scales performance, what new risks show up for you – things like bias, brand safety, or creative fatigue and how are you managing them?

Across many industries, there remains a significant gap in understanding AI's workings, capabilities, and limitations. Many people are unsure about the data needed and how to instruct AI to achieve desired results.

Early-career professionals can lack clarity on ‘what good looks like’, making it difficult for them to evaluate AI outputs. As senior leaders, we must guide and coach the new generation on quality results to leverage AI effectively. Those entering our industry have the advantage of expert knowledge at their fingertips. For instance, AI can simplify complex topics like CAPI (Conversions API).

Another challenge is the expectation gap. People often assume AI will provide perfect answers instantly and are disappointed when it doesn't. Training teams to use AI as an augmentative tool, requiring proper shaping, review, and evaluation, is crucial. The next generation of skills will focus on verification, critical thinking, attention to detail, and understanding what makes for high-quality output.

Despite advancements, quality and safety risks remain, such as AI hallucinations, biases, brand safety concerns, and producing what I call ‘confident nonsense’. This is where the importance of effectively prompting AI cannot be understated. Bias and brand safety are managed through strict guardrails, human oversight, effective prompting skills, adherence to evidence-based standards, and rigorous rules regarding sensitive data.

How do you balance the short-term gains of AI automation with the long-term need to build brand equity, deliver performance, improve productivity?

Validating a technology or skill in these terms can be challenging. While you can measure the time it takes to deliver a brief response and calculate the FTE, quantifying the enhanced quality of work provided by AI is more complex. Clients are experiencing better, faster, and more accurate responses from our teams.

Remember, don't sacrifice brand integrity for speed. Automation manages throughput, but people maintain meaning, brand voice, and long-term memory. Convert time saved into advantages like more testing, better creative direction, deeper insights, stronger measurement, and sharper recommendations. Aim for both productivity uplift and better outcomes, balancing both aspects.

Looking three to five years ahead, what capabilities do you hope AI will unlock that feel out of reach today?

My dream is that we will manage a small team of agents dedicated to specific tasks. This will grow to agents capable of handling complex tasks that typically take several days by understanding the context, identifying problems, devising solutions, and executing the output. These will then amalgamate into role-specific assistants with built-in context, standards, and workflows. These assistants will not only suggest actions but also execute, check, and improve within defined limits.

The true value will lie in our knowledge, making conversational interfaces the norm. We will be able to articulate our thoughts and see them turned into action seamlessly.

If AI could instantly amplify one human skill across your entire organisation, which skill would you choose and why?

Accelerated learning. The greatest advantage today is the ability to quickly understand, adapt, and level up. With an expert on demand for whatever you want to learn, however you want to learn it. I use it relentlessly.

What's one piece of advice you could give to those reading to help them on their AI literacy journey?

It would be to stop using AI like a vending machine.

Treat it like a strategic teammate: give clear context, constraints, examples, and a definition of ‘good’. Too many people stop here – don't. Iterate and evaluate, ask "How can this be better?". That is how you get great work, not just fast work.

Ultimate Amplifier: how Microsoft can help

Further insights from Adam Goodman, Director of AI in Advertising, Microsoft Advertising APAC.

AI is reshaping the marketing and advertising landscape faster than any previous technology shift, moving discovery away from traditional search toward conversational, multimodal, and agent‑driven experiences. Marketers are increasingly seeking clarity on how to participate in AI environments, and Microsoft’s offerings from Copilot’s conversational ad experiences to next‑generation tools like Brand Agents provide a practical pathway forward.

Copilot podcasts make micro‑learning effortless for marketers by turning complex topics into short, high‑impact audio explainers that fit naturally into the workday. Each episode can break down a single concept (like CAPI, retail media, or agentic optimisation) into a clear, digestible narrative that someone can learn - even while commuting or between meetings. It’s a fast way to stay sharp and something I personally have been using in recent months.

You can find this in your Library section in Copilot: https://copilot.microsoft.com/library

In this article we highlighted some of the ways Kellyn and agencies like Zenith are leveraging, learning and focussing on “AI for the individual and AI at scale”. Here are a few additional resources to help you consider how Microsoft, Copilot and Microsoft Advertising can help you on your own journey . . . to be your Ultimate Amplifier.

AI Marketer’s Guide (2025 edition): Practical blueprint for adapting to AI‑powered search, content structuring, and measurement in an AI‑first world. A marketer's guide to chatbots and agents: from generative AI to ROI | Microsoft Advertising

Generative AI and Automation Advertising Learning Path: Discover how artificial intelligence (AI) and automation can enhance your Microsoft Advertising campaigns. Learn the benefits and features of various AI powered tools and solutions that can help you meet and exceed your advertising goals. Learning Lab | Microsoft Advertising

What’s next for the All in on AI series? We’ll be exploring topics from AEO & GEO, creativity, productivity, the future of work, agentic commerce and more. Don’t miss next month’s leadership spotlight - coming soon.