What’s the missing ingredient in the conversations around AI projects? Hint: It’s not just about the tech. Amidst the excitement and hype, there’s a crucial voice often overlooked: the CFO. The financial implications of AI adoption are vast, making CFOs pivotal players in ensuring these projects succeed.
I loved the perspectives shared by a recent Gartner video—’How CFOs should navigate AI’: https://www.youtube.com/watch?v=y268jrtjako
AI projects involve significant costs that CFOs must navigate. Here are four types of costs associated with AI implementations:
1. Initial Rollout Costs: This includes infrastructure, software, new talent acquisition, and implementation costs. For example, investing in advanced GPUs and cloud services, and onboarding data scientists.
2. Ongoing Costs: Maintaining models, ensuring compliance, continuous data maintenance, environmental costs of running large models, and usage costs (e.g., per query charges).
3. Experimentation Costs: Costs associated with experimenting and potential failures, such as low adoption rates or selecting the wrong use cases. For example, piloting AI models that may not meet business expectations, requiring iterative development and adjustments.
4. Sunk Costs: Costs incurred that cannot be recovered, often due to abandoned projects or failed experiments. For instance, investing in an AI tool that does not integrate well with existing systems or exploring a use case that turns out to be non-viable.
Despite the potential benefits, there are also four AI stalls that CFOs need to manage to ensure successful AI integration:
A. Cost Overruns: Due to the newness of AI, cost estimates can be significantly off, sometimes by as much as 500-1000%.
B. Misuse in Decision-Making: AI’s role in decision-making needs to be carefully managed. Moving too fast can lead to poor decisions.
C. Loss of External Trust: As AI interacts with customers and stakeholders, maintaining trust is crucial. These needs to be governance around AI investments to ensure ethical, compliant, and transparent AI operations.
D. Rigid Mindsets: The human factor is critical. Resistance to change can stall AI initiatives. There is a need to invest in changing mindsets, fostering a culture that embraces AI as a tool to enhance rather than replace human roles.
In conclusion, while the technical and operational benefits of AI are evident, it is the CFO’s strategic management of these costs and potential stalls that will determine the true success of AI projects.
