[Brand](https://sustainablemarketing.com.au/blog/brand) [Digital](https://sustainablemarketing.com.au/blog/digital) [Marketing](https://sustainablemarketing.com.au/blog/marketing) [Public Relations](https://sustainablemarketing.com.au/blog/public-relations) # I Wasn't Trying to Build an AI Operating System. I Was Simply Trying to Make AI Work Better. Posted 20 July 2026 by Maria Anderson ## I Wasn't Trying to Build an AI Operating System. I Was Simply Trying to Make AI Work Better. ### **Why better AI wasn't the answer. Better systems were.** Artificial intelligence has become part of almost every business conversation. Some businesses are using it to write content. Others are automating customer service, creating proposals or building custom AI assistants. Every week there seems to be another tool promising to save time or transform the way we work. Yet when I speak to business owners, I hear a similar story. *"Sometimes it's brilliant. Other times it's completely off the mark."* I understood exactly what they meant because I'd experienced the same thing. When I first started exploring AI, I wasn't trying to build an AI operating system. I simply wanted to understand whether AI could help us work more efficiently and support better outcomes for our clients. Looking back, I realise I approached AI the same way I've approached most business challenges throughout my career. I'm a systems thinker. When something feels inconsistent, I don't immediately look for a better tool. I start asking different questions. *Why does it work well sometimes and not others?* *What's missing?* *What does the system around it look like?* At the time, I didn't think I was building an AI operating system. I was simply trying to make AI work more consistently for our business. Only later did I realise that's exactly what I had been building. How my thinking evolved from AI tools to an AI operating system ### **The first lesson: AI only knows what you teach it** Like many people, I started with ChatGPT. At first, the responses were impressive. Until they weren't. I could ask it to write a proposal or develop a marketing strategy and it would produce something that looked polished. The grammar was good. The structure was logical. But something was missing. It didn't sound like Sustainable Marketing Services. It didn't reflect how we think about strategy. It didn't understand why we often recommend a slower, more sustainable approach rather than chasing every marketing trend. It didn't know our clients. It didn't know our decision-making process. Most importantly, it didn't understand the judgement behind our recommendations. At first, I assumed I simply needed to become better at prompting. Like many people, I spent time experimenting with different prompt structures, refining instructions and trying to get more consistent responses. Sometimes it worked. Sometimes it didn't. The inconsistency wasn't frustrating because AI wasn't intelligent. It was frustrating because I knew our business had decades of knowledge that AI couldn't possibly see. The problem wasn't the tool. The problem was the context. That was my first real turning point. ### **Better prompts weren't enough** Around this time, I became part of the [AI Society](https://aisociety.socialmediaexaminer.com/?smesite=aisociety25-site-navbar#top) through Social Media Examiner. One of the sessions featured a guest from [AI Queens](https://www.aiqueens.com), who generously shared her Chief Marketing Officer GPT. Like many people in the session, I was fascinated by what it could do. But what interested me most wasn't the technology. It was *why* it worked so well. For the first time, I could see that the quality of the responses wasn't coming from clever prompting alone. It came from the strategic thinking that had been built into the assistant. Expert frameworks. Marketing principles. Structured reasoning. That was a lightbulb moment. I realised AI became significantly more valuable when it had access to proven strategic frameworks, not just information. That insight changed the way I started building our own AI assistants. More importantly, it shifted my thinking from prompts to strategic frameworks. I realised AI becomes far more valuable when it can draw on proven ways of thinking, not just information. That insight eventually led me to build Expert Cards into many of our Custom GPTs and Claude Skills, giving them access to structured marketing, business and communication expertise rather than asking them to start from scratch each time. Instead of simply giving them facts about our business, I began incorporating the strategic frameworks that shape how we think and make decisions. Around the same time, I started using Expert Cards to capture the thinking of respected marketing and business experts whose approaches aligned with our own. Those cards have since become a consistent part of many of our Custom GPTs and Claude Skills, helping them reason more strategically rather than simply generate content. Looking back, that was probably the first time I stopped thinking about AI as a tool and started thinking about it as part of a much bigger system. ### **Knowledge is only useful if it's captured** As my use of AI evolved, another challenge became obvious. Even the best strategic frameworks weren't enough if the AI didn't understand *our* business. It needed more than marketing knowledge. It needed to understand us. Around this time, I worked with [Loren Bartley](https://impactiv8.com.au/lorenqr/) to develop both my Personal DNA and the Business DNA for Sustainable Marketing Services. That process wasn't really about AI. It was about making the invisible visible. We documented how decisions are made. Why we recommend one approach over another. The values that shape our advice. How we communicate. What sustainable growth actually means in practice. The kinds of clients we do our best work with. The boundaries we won't compromise. Those weren't documents created for AI. They were created to help us better understand ourselves. The unexpected benefit was that they also became some of the most valuable knowledge sources for our AI ecosystem. Because once that thinking was documented, AI could begin working from the same foundations we use ourselves. It wasn't replacing our judgement. It was learning from it. ## **Guardrails don't limit AI. They build trust.** As our AI capability grew, another question became increasingly important. It wasn't just ... *what could AI do?* It was ... *what should it do?* That wasn't a new question for Sustainable Marketing Services. Our business had always been guided by clear values, ethical marketing practices and a commitment to making sustainable growth achievable. We already had boundaries around the advice we give, the clients we work with and the way we approach marketing. The difference was learning how to extend those same principles into AI. One of the biggest influences on that thinking came through [Dr Nici Sweaney's work](https://aiherway.com.au) on AI operating systems and governance. She helped me see that if AI was going to become part of how we worked, it also needed to understand our guardrails. Not just our knowledge. Not just our processes. Our judgement. That meant documenting how we approach ethical decisions, where human oversight is essential, what quality looks like, how we protect client confidentiality, how we communicate respectfully and inclusively, and where AI should never replace professional experience. Our guardrails didn't change. We simply learnt how to embed them into our AI systems. Looking back, I think that's one of the biggest misconceptions about AI. People often assume governance slows innovation. My experience has been the opposite. Clear guardrails create confidence. They allow you to innovate because everyone, including your AI, understands the boundaries. ## **One assistant couldn't do everything** At first, I assumed one well trained assistant could do everything. The more our AI capability grew, the more I realised that wasn't how healthy organisations work. We don't expect one person to lead strategy, marketing, finance and operations. Each role exists for a reason. From One AI Assistant to an AI Team So rather than building one increasingly complicated assistant, we began designing specialists with clearly defined responsibilities, boundaries and handoffs. Blair became our AI Chief Marketing Officer, helping develop marketing strategy, campaigns and content while staying aligned with our brand, positioning and marketing system. Sage became our AI Business Strategist, supporting decisions around pricing, profitability, capacity, offers and sustainable business growth. Cleo became our AI Chief Financial Officer companion, helping us understand financial performance, cash flow and profitability while encouraging calm, confident decision making. Each assistant has a clearly defined purpose. Each has boundaries. Each knows when another specialist is better suited to the task. That wasn't just an AI design decision. It reflected how healthy organisations work. ## **Strategy before execution** As more specialist assistants emerged, I realised they all needed something in common. A shared understanding of the business. That's where Aria evolved. Aria isn't another content assistant. She isn't another marketing GPT. She acts as our strategic intelligence and governance layer. Before work moves to a specialist, Aria helps test whether it aligns with our business strategy, positioning, ideal clients, guardrails, values and long-term direction. Sometimes the recommendation is to proceed. Sometimes it's to refine the idea. Sometimes it's to redirect it entirely. That strategic thinking happens before execution begins. It's another example of something we've always believed at SMS. Strategy before execution. Clarity before activity. AI simply gave us another way to apply that philosophy consistently. ## **An operating system should keep evolving** One thing I've learnt is that an AI operating system is never finished. Every new framework, every refinement to our business knowledge and every documented process makes it stronger. Most recently, I've started using a Standard Operating Procedure Creator developed by [Kate Vandervoort](https://aisuccesslab.com/home). Again, what interested me wasn't simply generating SOPs faster. It was that Kate had designed it to become part of an AI operating system. That meant we could integrate it into ours, creating SOPs that align with our business, our standards and the way we work. Like every other component, it doesn't replace our thinking. It strengthens the system around it. That's probably the biggest lesson I've learnt over the past few years. Better AI wasn't the goal. Building a better system was. ## **So what is an AI operating system?** People sometimes ask me what I actually mean by an AI operating system. For me, it's much more than a collection of AI tools. What Makes an AI Operating System? It's a connected system that combines: **Knowledge** The documented expertise that makes your business unique, including your strategy, positioning, methodology, operating principles, customer insights and intellectual property. **Guardrails** The governance that ensures AI operates consistently with your values, ethics, quality standards, decision making and human oversight. **Specialists** Clearly defined AI assistants, Claude Skills and other capabilities, each with a specific role, responsibilities and boundaries, working together rather than trying to do everything. Technology will continue to change. These foundations shouldn't. ## **Looking back** When people ask whether AI has changed the way I work, my answer is yes. But probably not in the way they expect. The biggest change hasn't been learning new technology. It's been understanding our own business more deeply. Building an AI operating system forced us to document knowledge that had lived in people's heads for years. It challenged us to articulate our values. To clarify how we make decisions. To define roles. To establish guardrails. To capture expertise that might otherwise have remained implicit. In many ways, AI didn't transform our business. It helped us understand it better. If you're experimenting with AI and finding the results inconsistent, my advice isn't to look for a better prompt or another tool. Start by understanding your business. Capture your knowledge. Define your guardrails. Clarify your roles. Then build a system that brings those pieces together. Because sustainable results rarely come from individual tools. They come from well-designed systems. And perhaps that's the biggest lesson AI has taught me. Looking back, I don't think I built an AI operating system because I was interested in AI. I built it because I'm interested in helping businesses grow in a way that's structured, sustainable and achievable. AI simply became another part of that system. The technology will continue to evolve. New tools will emerge. Better models will be released. What won't change is the importance of strategy, knowledge, guardrails and human judgement. Because sustainable growth has never been about finding the next shiny tool. It's about building systems that help good people make better decisions, deliver consistently and keep improving over time. **For me, that's what an AI operating system is really about.** **Key Takeaways** If you only remember five things from this article, let them be these: - Better AI starts with better business knowledge, not better prompts. - Strategic frameworks and documented expertise help AI reason more effectively. - Guardrails and ethics don't restrict AI, they create consistency, trust and confidence. - Specialist AI assistants with clear roles outperform a single assistant trying to do everything. - An AI operating system should evolve alongside your business, strengthening how you think and work rather than replacing human judgement. If you're experimenting with AI but finding the results inconsistent, the answer may not be another tool. It may be building the system around it. At Sustainable Marketing Services, we're exploring how strategy, knowledge, guardrails and AI can work together to make sustainable growth achievable. If you're exploring how AI could support your business, don't start with the tools. Start with the system. If you'd like to explore what that could look like for your business, [let's have a conversation](https://sustainablemarketing.com.au/contact). **Frequently asked questions** An AI operating system is a structured approach to using AI within a business. It combines documented knowledge, governance, guardrails, specialist AI roles and human oversight so AI supports consistent, strategic decision making rather than operating in isolation. ChatGPT is a powerful AI tool. An AI operating system is the broader framework around it, including your business knowledge, strategic frameworks, governance, guardrails and specialist AI assistants that work together to achieve consistent outcomes. Guardrails help ensure AI aligns with your business values, ethical standards, quality expectations and strategic goals. They also define where human judgement remains essential, building trust and reducing risk. Not necessarily. The right approach depends on the size and complexity of your business. However, clearly defining the role of each AI capability often produces better outcomes than expecting one assistant to do everything. Start by documenting your knowledge. Capture your strategy, customer insights, processes and decision making before investing heavily in tools. AI becomes significantly more valuable when it understands how your business thinks and operates. No. AI can support research, analysis and execution, but strategy, judgement, ethics and relationship building remain human responsibilities. The most effective AI operating systems strengthen human expertise rather than replace it. Regularly. As your business evolves, your knowledge, operating procedures, guardrails and AI capabilities should evolve with it. An effective AI operating system is designed for continuous improvement. About the Author [**Maria Anderson**](https://sustainablemarketing.com.au/maria-anderson-brisbane-marketing-consultant) **is the Founder and Managing Director of Sustainable Marketing Services, where she helps growth-focused businesses make sustainable growth achievable through clear strategy, structured marketing systems, and practical leadership.** With more than 25 years' experience in marketing, Maria specialises in turning fragmented marketing into a connected, measurable system that gives business owners confidence in where to focus, what to prioritise, and how to grow sustainably. A Certified Practising Marketer (Australian Marketing Institute), Maria holds a Bachelor of Business (Marketing and Public Relations) together with Diplomas in Sustainability and Financial Services. She combines strategic thinking with practical implementation, helping businesses simplify marketing, strengthen their positioning, and achieve measurable long-term growth. Maria is also the creator of the Sustainable Growth System—a strategy-led approach that connects strategy, leadership, execution, and optimisation into one practical marketing framework. She believes great marketing should never feel overwhelming. It should be clear, manageable, measurable, and achievable.