Generative-AI Capabilities, Promise, and Practice
In this multi-part series, we examine the potential role of generative artificial intelligence (AI) in accountancy. First, we look at the capabilities of generative AI, the needs of the accounting industry, and how generative AI capabilities can help. We review the risks involved in the use of artificial intelligence in general and in accounting. Then we consider the industry standards and regulatory bodies and their policies and recommendations for the use of AI. We examine the various ways of implementing a generative AI accounting solution involving large language models: manual, custom API, and automated, public, commercial APIs. Then, we look at the implementation of AI on private, accounting databases before we conclude.
Generative AI Capabilities
Generative AI produces new content through the use of algorithms, large language models (LLM,) and neural networks (computer architecture in which microprocessors are connected in a manner suggestive of connections between neurons, which, can learn by trial and error.) Machine learning involving algorithms is combined with third-party creative works to produce the new content. The internet is one of its repositories for third-party text (scholarly works, guides, encyclopedias, laws, codes, regulations, literature (novels, poems,) &c.,) music (score,) audio, video, &c.
Generative AI can accomplish analysis tasks and generate an original set of data and/or business processes. It can be used to boost the work of writers, graphic designers, artists, and musicians. It is useful in describing products, suggesting design variations, or helping examine different concepts; and is capable of natural language processing and generation, translation, and can be used as groundwork for other tasks. These capabilities can be adapted to the accounting field.
The Promise And Practice of Generative AI in Accounting
The automation of routine tasks, such as data entry, table formatting (including data preparation: data extraction, cleaning, integration (combining data from different sources for a unified view,) and transformation (conversion of data from one format to another),) invoice processing, bookkeeping, and payroll are some accounting tasks within the competence of generative AI. Further, though, it can produce summaries, enhance the speed of financial analysis and decision-making, and make recommendations and predictions (predictive data analytics.) It can improve the accuracy and reliability of financial reports by reducing errors and minimizing human biases. It can provide real-time insights and alerts on financial performance, risks, and opportunities. It can substantially assist in the audit process inclusive of detecting fraud and breaches of regulatory compliance and organizational policy. It can facilitate income tax preparation.
Skilled Accountants Shortages
Consequent to these capabilities, as there are shortages of skilled accountantss, some pundits, alarmed at the shortages of skilled personnel in the professional services industries, a shortage that is forecasted to grow in the next decade, see generative AI as a means to fill the gap. TOA (The Outsourced Accountant,) an Australian professional services firm, observed the shortages:
In 2022, the global skills shortage reached a 16-year high, following a 6% year-on-year rise. In 2012, 34% of employers struggled to find the talent they needed. This has doubled in a decade to 75% in 2022. If this trend continues, as it’s predicted, we won’t have to wait another decade to be faced with a world that looks very different according to forecasts.
- The number of occupations experiencing shortages in Australia rose from 153 to 286 in a single year, 2021 to 2022
- Over 90% of Australian and New Zealand businesses expect to be hit by the shortage before mid-2023.
- The US is facing $8.5 trillion in unrealised [sic] revenue and may be surpassed by India as the world’s main tech giant.
And again, in the same place, in 2021, the accounting skill shortage in the US was exacerbated by experienced staff turnover:
The 2021 financial reporting season saw corporate finance teams and their outside auditors having to deliver accurate financial statements to investors with fewer staff on hand. The Public Company Accounting Oversight Board is concerned that the turnover of experienced staff is an “emerging audit risk”.
The lead author of a major annual study on the financial performance of accounting firms in the US, has said that “Without exception, firms are trying to understand how they can deal with the labor shortage in the profession”. He warns the issue will only get worse before it gets better and advises firms to think outside the box about how they’ll get work out the door.
–Source: “What You Need to Know Now to Survive the Skills Shortage.” TOA Global. 2024, https://toaglobal.com/au/blog/accounting-firms-skills-shortage-crisis/
Generative AI to The Rescue
As mentioned above, some pundits see a role for artificial intelligence in completing routine tasks.
AI tools might alleviate the shortage of skilled workers by automating routine tasks and augmenting human capabilities. This has already been happening. Take cybersecurity, which is currently short about 3.4 million skilled workers globally, according to a 2022 study. To cope, organizations are turning to AI to automate the detection and response to cyber threats, freeing up cybersecurity professionals to focus on innovation.
–Source: Schwarz, John. “Can AI Help Solve The Workforce Skills Gap?” Forbes, Apr 6, 2023, https://www.forbes.com/sites/forbestechcouncil/2023/04/06/can-ai-help-solve-the-workforce-skills-gap/?sh=77d37152134f
The range of generative AI capabilities, from routine to sophisticated analyses and decision-making, lend themselves to use in professional services firms like accounting practice. As a consequence, the emergence of these generative AI capabilities is timely to mitigate the current shortages of skilled professionals in professional services industries, inclusive of the accounting industry, and dampen the projected worsening, within a decade, of the shortage for the skilled professional services industry. Next time we will look at the risk associated with the use of generative AI for accounting.
–Richard Thomas
Next, Part II
