AI Readiness Series

Introduction

The pace of change and innovation, along with the accessibility of ChatGPT and other generative AI tools, has galvanised almost every organisation. The Australian Artificial Intelligence (AI) market is growing significantly and is projected to be worth US$315 billion by 2028, according to The Commonwealth Scientific and Industrial Research Organisation (CSIRO). At CSP, we are passionate about ensuring these opportunities presented by AI are balanced by considering the associated risks. We care about Australian businesses, and this has prompted us to write a series of blogs that we will release in the coming weeks. In them, we discuss six key stages of AI readiness when using Microsoft solutions and services.

People are becoming increasingly familiar with Microsoft’s flagship AI offering known as Copilot and the appetite for adoption is gathering pace. Industries are already being augmented by AI, Intelligent Automation, Digital Humans, and multi-agent systems, and we are really only at the early stages of this revolution!

Mindset and Responsibility

Over the last twelve months we have been working closely with the best thought leaders in AI enablement across South Australia. See this recent post from Adelaide start-up Digital Human Assistants (DHA) on AI knowledge certainty and what it means for decision-makers. This got us thinking about the philosophical underpinnings of AI, and how it can shape how we approach readiness and deployment of AI solutions.

Following in the tradition of Karl Popper, we view the world as a series of continuous problems and solutions, and consider knowledge (including when AI generated) as inherently provisional. Since CSP’s founding, we have always advocated for a mindset that embraces uncertainty and prioritizes continuous improvement in order to build resilience and correct errors. We see the development of an AI readiness roadmap as the most effective path to successful AI adoption.

AI Shared Responsibility Model

Many risks can be mitigated by appropriate and responsible use, which is why Microsoft explicitly requires customers to comply with the Acceptable Use Policy within the Product Terms, as applicable, or the Azure OpenAI Code of Conduct. Microsoft also encourages customers to train their users in understanding the limitations and fallibility of AI.

The following diagram illustrates the areas of responsibility between you and Microsoft according to the type of deployment. As you’re performing risk assessments, keep the shared responsibility model in mind. Copilot for Microsoft 365 is a Software as a Service (SaaS) solution and we’ve outlined some important areas for consideration and included them in our custom roadmap.

The Future

The initial step in any AI business strategy is to identify the problem you’re aiming to solve. This will inform which use cases you’ll select, how you’ll prioritize them, how you’ll measure success, and how you’ll invest in AI for maximum impact. Clear business goals for AI promote alignment of projects to strategic objectives—such as efficiency and revenue generation—things that the C-suite and board care about and are more likely to generate visibility and funding. Microsoft’s AI business strategy underscores five stages which we’ve included below and our overall aim of this series is get you geared up for the first two from a data security perspective.

In the coming months, we’ll be releasing a series of blogs that delve into six key areas we believe really matter. We hope our AI readiness roadmap will serve as a ‘north star’ as you navigate the exploration and planning phases of your AI journey.

We will also be giving a free 45 minute Teams call (which you can book here), if you prefer to discuss the Microsoft digital estate face to face.