In the past year, generative AI has captured the public’s imagination and is beginning to transform the way work is done.
As finance leaders look to capitalize on the latest developments in both traditional (i.e., data science based) and generative AI, many are examining their current business processes to explore where AI can deliver value embedded in existing workflows.
These technologies will increase productivity, improve decisions, and reduce costs. But, like other emergent technologies, there are numerous considerations before finance leaders embark on a journey to incorporate these technologies within their current processes and workflows.
A successful AI strategy hinges on measurable results as well as employee adoption. It’s important at the outset for finance leaders to define key performance indicators that align with their business goals. While it can be tempting to immediately jump into setting and tracking quantitative goals such as improvements to overall productivity and increased forecast accuracy, there are critical “soft metrics” to regularly set and assess as well.
These include tracking the levels of comfort and confidence that employees feel in using these technologies and the levels of acceptance and usage across specific departments and the organization at large. Tracking these types of soft metrics at the outset helps to establish a more sustainable AI strategy that is grounded in employee buy-in and advocacy.
When it comes to implementing AI, finance leaders need to think big but start small and approach new projects with a sense of “radical practicality” before reaching for shiny objects or diving head-first into massive AI deployments.
Don’t let employees take AI matters into their own hands, as there are significant data privacy and security risks associated with using consumer-focused large language models (LLMs) in a business setting. Organizations will be better served when finance leaders enable access to AI technologies within the context of enterprise resource planning (ERP) or enterprise performance management (EPM) applications.
This helps to prevent leakage of sensitive data, while also safeguarding against issues like hallucination. Finance leaders must do their due diligence to find application vendors and/or LLM providers that ensure that corporate data is not shared across organizations and add a layer of domain expertise in how the models are trained and deployed.
Finance leaders need to acknowledge that there are real concerns from employees regarding how these technologies will impact their roles over time. These concerns are valid and must be acknowledged. AI and generative AI technologies are here to augment and support specific roles by streamlining and automating repetitive tasks to allow for increased productivity and innovation.
There will be certain employees who are already ahead of the curve and are well suited to be early AI adopters. Finance leaders should look to showcase these employees as examples and as advocates to instill a sense of confidence in other team members and dispel some of the fears that are associated with automation. Finance leaders should also look to establish centers of excellence where peer groups can learn from one another. Additionally, education programs need to be in place for employees to learn more about data and AI literacy.
Many finance teams are already tapping traditional AI to assist with detecting patterns, analyzing anomalies, and providing data-driven recommendations. As we look towards this year, we’ll begin to see more finance teams integrate generative AI capabilities to summarize key financial documents, generate drafts of reports, explain data-driven insights, generate accounting and audit notes, and more.
The ongoing evolution of AI and generative AI will unlock new possibilities and help optimize and automate workflows within the finance function. The leaders who embrace AI now and align their initiatives to broader business operations and strategies will reap greater benefits than those who sit on the sidelines.
LeackStat 2024
2024 © Leackstat. All rights reserved