We’ve reached the point of talking about "traditional" AI capabilities—ones you’re likely familiar with and may use already. Some Finance organizations have broadly used capabilities like robotic process automation (RPA) and natural language processing (NLP) to automate tasks, streamline individual tasks, and uncover patterns and correlations. These tools have shown they can automate routine activities and reduce human workloads, but they cannot replace human workers when it comes to addressing complexity, ambiguity, or surprises.

Machine learning is a data analysis method that uses algorithms and historical data to identify patterns and make predictions. The “learning” in its name comes from real-world contexts, not programmed rules. High-volume finance activities that require judgment-based decisions can benefit from ML—in chatbot interactions, predictive forecasting, and tax compliance, for example. But ML can’t work past anomalies or substandard data the way humans can. At least, not yet. For those curious about generative AI, ML could be a start, with similar use cases, but implemented generally faster and less expensively (for now).

Behind the current buzz is a powerful new kind of AI in which machines can create new content that mimics human work—text, code, voices, images, videos, and more. For Finance, the technology’s full potential may still be in the (near) future. For now, though, generative AI can function like an always-on, automated analyst: It can prepare a budget or write a report, but you wouldn’t take its work to the board without checking it yourself. Still, because it can "team up" with humans so quickly and efficiently, generative AI appears likely to spur dramatic changes in finance work in the years ahead.