Use AI to amplify your expertise, accelerate decisions, and strengthen your leadership impact.
As financial leaders, we’re often seen as the guardians of certainty in a world of constant change. We’re expected to provide clear, data-driven answers that guide the business forward. Yet, the conversation around Artificial Intelligence often feels like the opposite of certainty. It’s a swirl of hype, anxiety, and vague promises that can leave even the most seasoned executive wondering what’s real and what matters.
After more than 40 years in technology consulting, working at KPMG, BearingPoint, and Deloitte, I’ve seen countless technology cycles come and go. What I see now with AI is different. This isn’t just another software upgrade; it’s a fundamental shift in how we can approach our work. AI is poised to become your greatest asset, a collaborator that can elevate the execution of everything you do.
The crucial point isn’t about AI replacing financial expertise—it’s about amplifying it. This article is about how you, as a finance leader, can harness AI responsibly. We’ll move past the abstract and into the practical, exploring how to make AI a trusted partner for analysis, strategy, and decision-making. My goal is to give you a clear framework for turning this powerful technology into a tangible advantage.
It Starts with a Conversation, not a Command
The most common mistake I see is treating generative AI like a search engine. We ask it a question and expect a perfect, final answer. But the real power of AI is unlocked when you treat it like a new team member. You wouldn’t expect a junior analyst to deliver a board-ready report without any context, and the same principle applies here.
The quality of the output you receive is directly influenced by the quality of your input. Before you ask AI to perform a task, you need to set the stage.
- Explain your role: Tell the AI who you are. “You are assisting a CFO of a mid-sized manufacturing company.”
- Define its role: Give it a persona. I often use a prompt that creates a virtual team, assigning roles like a strategy lead, a research analyst, and a QA editor.
- Provide context: Give it information about your organization, your goals, and the specific project.
- Set boundaries: Establish the parameters of the interaction. Define the tone, the scope, and the desired format for the output.
By structuring your interaction this way, you transform AI from a simple pattern-recognition tool into a collaborator that understands your intent.
From Blank Page Paralysis to Strategic First Draft
How many times have you stared at a blank screen, whether for a board presentation, an investor update, or a new financial model? That initial step is often the hardest. This is one of the most immediate and practical applications of AI. Let it take the initial pass.
AI is fantastic at overcoming inertia. Because it doesn’t have the preconceived notions or mental blocks that we do, it can generate a comprehensive starting point in seconds. Think of it as brainstorming on steroids. It might produce ideas or structures you hadn’t considered.
The key is to remember that you are still the expert. You don’t take the output verbatim. You evaluate it, critique it, and refine it with your own experience and judgment. Use AI to handle the initial 70-80% of the work—the routine analysis, the data structuring, the narrative outline. This frees you up to focus on the most valuable 20%: the strategic insights, the nuanced storytelling, and the final executive polish.
Choosing Your Model: From Public Playgrounds to Enterprise Fortresses
Not all AI is created equal, and for finance professionals, security is non-negotiable. It’s crucial to understand the different models available and where they fit.
- Public Models: Tools like ChatGPT Consumer or Perplexity offer a low-cost, accessible way to start exploring AI’s capabilities for general, non-sensitive tasks. They are excellent for everyday use, research, and content creation that doesn’t involve proprietary information. Many of the models allow you to turn off a feature to use your data for model training.
- Enterprise Models: When dealing with sensitive company data, you need a solution with robust security and compliance features. Enterprise-grade platforms such as ChatGPT Enterprise or Claude Enterprise offer SOC 2 compliance, GDPR protections, and a secure environment for your data. They ensure that your information isn’t used to train the public model. Enterprise models are delivered with your data feature turned off.
- Self-Hosted Models: For maximum security, some organizations opt for self-hosted, private AI models. In this setup, your data never leaves your own servers. This offers the ultimate control but requires significant technical resources to implement and maintain. Self-hosted AI models are designed to process data locally without sending user inputs or outputs back to vendors for training purposes.
For a finance executive, the takeaway is clear: start with public models for low-risk tasks to build your skills. But for any work involving financial statements, HR data, or corporate strategy, you must operate within a secure enterprise or self-hosted environment.
Making AI Your Institutional Memory
One of the most powerful but overlooked aspects of modern AI tools is their ability to build and retain context. In many platforms, you can create separate projects or folders for different clients, initiatives, or departments.
Within each project, you can load specific files—like your org chart, financial statements, or IT inventory. You can also provide “custom instructions” that define your communication style and project-specific goals. The AI then builds a “memory” within that context, ensuring its responses are consistently relevant and informed by the documents you’ve provided.
Imagine having an assistant who has already read every financial report from the last five years and can instantly reference a specific data point from a board meeting two years ago. That’s the power you can build. It reduces manual friction and ensures that the AI’s analytical support is always grounded in your company’s reality.
Your First Step Forward
Getting started doesn’t have to be complicated. If a task is repetitive, complex, high-volume, error-prone, or requires high accuracy it’s a prime candidate for AI assistance.
Here are a few questions I encourage you to ask your team:
- Where do we spend the most time on manual data aggregation or reconciliation?
- Which reports take us the longest to draft each month or quarter?
- What strategic questions could we explore if we had an extra analyst dedicated to research?
The answers will point you directly to your best opportunities for integrating AI. A simple first step is to take a recent, non-confidential report you’ve written and ask an AI tool to summarize the key findings or draft an alternative executive summary. Compare its output to yours. This small exercise will begin to build your intuition for how to work with these tools effectively.
The landscape of finance is evolving, and our ability to integrate technology like AI will define our impact as leaders. It’s not about letting a machine take over; it’s about leveraging a powerful new collaborator to do our best work.
If you’d like to explore these ideas in more detail, you can watch my full session with The Financial Executives Networking Group here: https://youtu.be/Pfyfs_JNSNc. I’m always interested in how other financial leaders are approaching this—feel free to connect with me on LinkedIn or visit my website https://thezasogroup.com.
