To say artificial intelligence is all around us would be an understatement. AI has become close to a utility. As I write, I am under the watchful gaze of the AI that is incorporated into the tool I am using to write this and I am sure, given the chance, the tool would have a thing or two to say about what is being written.
What makes AI so different from everything before from a technology perspective is how widely it is being integrated into everything else that already exists. AI has survived where other technological innovations such as blockchain, cryptocurrency, big data or anything else has only briefly shone before being relegated to a specialized nook of application. AI, it seems, will have an increasingly longer shadow into the future simply because it has become so easy to integrate into anything else that exists, from ERP solutions to chatbots to browsers to tools that are used for making notes. The promised land offers increased efficiency, improved decision-making, enhanced customer experiences and new revenue streams driving enterprises to adopt AI.
This ease and interest of integration is the game-changer and one that has implications for all of us in our personal and professional lives.
At the upcoming ISACA Virtual Conference, 18–20 February 2025, I am speaking about how enterprises can leverage AI for innovation, why enterprises need to focus on this and how they can do it. A simple search will tell you how many technology adoption projects are abandoned for poor or no return on investment. While ROI is an easy metric to look at, it is hard to calculate when it comes to technologies like AI primarily because AI is not one technology or a technology monolith, but an amalgam of technologies. And let’s not forget about data, an enterprise-specific component that needs to be included for everything to work effectively.
AI adoption by enterprises is afflicted by other challenges as well. Enterprises often tend to adopt AI because of external pressures rather than because they have a clear, well-articulated strategy for why AI is required and what they intend to achieve with it. The lack of strategy is exacerbated by, among other things, the lack of data or quality data. Often, poor data leads to poor outputs, and technology is usually blamed for the failures. Skills are a perennial issue and, at least in the foreseeable future, skill shortages promise to hobble AI adoption in enterprises. On the surface, these may seem like issues that could be resolved easily but considering the dynamic interactions AI and related technologies have with other parts of the enterprise, this can be challenging. Considering what AI is capable of and the outcomes possible, ethical considerations abound and need careful handling for AI initiatives to be successful.
Of course, it is not all doom and gloom. Enterprises should start by defining specific business goals that have to be achieved and problems to be solved that can be articulated as a business strategy for AI adoption. This AI strategy can be used to support and drive AI implementation within the enterprise. While this may seem like putting the cart before the horse, this is critical and will help define not just the expected result but also the path forward. Metrics that can be used to evaluate status periodically or at defined milestones can be derived, enabling course correction as required. It will also help if there is clarity on data and how controls around data are implemented, especially considering data quality and ethics.
People are a key building block for successful AI deployment in the enterprise, and we will discuss how we can equip people so that they can contribute to the successful adoption of AI at all layers right from conceptualization: design, development, deployment, governance, risk management, audit, etc. Risk in the context of AI needs a deep look, touching on risks at multiple levels right from the AI model itself, such as data, integration issues (tech debt), business process interfaces, supply chain management, incident management, and more. Long-term success will hinge of the adoption of what COBIT refers to as the goals cascade, where business strategy and goals previously established are connected to IT and other goals further downstream, enabling top-down guidance and bottom-up feedback facilitating effective implementation. We will discuss how to bring all these aspects together for a successful AI adoption and integration within the enterprise, so the promised ROI is realized.
Come to the virtual conference session for deeper insights and actionable takeaways, no matter your professional role or where you are in the AI supply chain. Looking forward to seeing you there!
