Where have all the accountants gone? It's a question more companies than ever are asking themselves as they try to address a growing talent crisis that has the potential to upend financial reporting, tax compliance and audit functions for businesses of every type.
More than 300,000 U.S. accountants and auditors left their jobs in the past two years— a 17% decline in employed accountants and auditors from a 2019 peak,
For adopters of new technology — most notably, generative artificial intelligence — those figures are not a problem; they are an opportunity. New technologies have promised a revolution in efficiency and streamlined processes that will allow accountants to do more with less. And importantly, to free them up to focus on adding more value to clients. But is that technology ready for primetime and can accounting professionals trust it to deliver the efficiency gains they need without compromising on accuracy?
Are accountants ready for AI?
Right now, the accountants at the center of the growing talent crisis are cautiously optimistic on the potential of generative AI. According to the recent
However, they also have some concerns. Some of the professionals' biggest fears around generative AI include compromised accuracy (25%), widespread job loss (19%), the demise of the profession altogether (17%) and concerns around data security (15%) and ethics (15%).
These responses make sense. New advances in AI-driven tax audit, provision and compliance technology are making it possible to transform accounting workflows by automating many of the repetitive, lower-value/higher volume tasks that consume time and add little in the way of job satisfaction or fulfillment. But not all AI solutions were created equal. When it comes to accounting, the difference between solutions that are fit for professional-grade work and those that could create more trouble than they are worth is all about the rigor with which those solutions were developed.
Put simply, before accountants can start reaping the benefits of generative AI, they need to be able to trust that the technology was developed responsibly. It is not enough to load a client's confidential financial information into the public version of ChatGPT and hope for the best. Professional-grade generative AI solutions that will be relied upon by accountants and other highly specialized workers need to be trained using comprehensive, authoritative data sets. And that process needs to be intermediated by human subject matter experts who understand the nuances and the context.
Separating the substance from the noise
Since generative AI burst into the mainstream and onto the innovation agendas of virtually every business in the world, a sort of Wild West phenomenon has started to take root. Companies have been launching proof of concept solutions, individual users have been experimenting with publicly available technology, and many examples of errors and abuses of the technology have come into the news spotlight. It is critical at this stage that the professional users who will benefit most from this technology understand the differences between generative AI solutions that were developed responsibly and those that are just putting on a good show.
The fact is, generative AI has the power to transform accounting in a fundamental and existential way. Scanning through reams of spreadsheets, scrutinizing terms and conditions, composing summaries and dashboards for clients — these are all areas where generative AI can have the biggest impact and they are the very areas where the accounting industry is struggling to keep pace and build engagement in its employee population. To extract those benefits, it will be critical to scrutinize the details and vet new technologies by asking questions like: "How was the model trained? What search queries and search results are used to produce the results? What role did human experts play in helping to tune the results and check for inaccuracies?"
We are in the midst of one of the most exciting periods of technological transformation since the advent of the internet. As we get deeper into the world of professional application of generative AI, it will be critical for AI to be trained on professional grade, trusted content and data, and that subject matter experts continue to wield a heavy hand throughout the development process.
