Finding a responsible approach to AI implementation

According to recent research, many accounting industry professionals believe artificial intelligence (AI) can improve practices — but only with the proper guardrails in place. Improving efficiency, reducing overhead costs, and providing better client and employee experiences are all attractive benefits in an era of talent constraints and changing customer expectations. The vital question is, how can accountants realize these benefits while managing the business risks associated with AI — particularly generative AI?

In January 2024, Accounting Today surveyed more than 200 accounting professionals to better understand the evolving role of AI in all its current iterations in the accounting industry. These survey findings suggest firms that identify and implement responsible AI tools can meaningfully improve their standing in the marketplace. Further, those firms that fail to do so are at risk of being left behind.

AI: A promising but worrying practice tool

Most accountants recognize the potential for AI tools to benefit their practices, especially opportunities to increase efficiency, automation, and productivity (see Figure 1). However, accountants tend to be more concerned about the rapid advancement of AI technology than their counterparts in peer verticals. In fact, 70% of accountants say AI is evolving too quickly, whereas a lower percentage of professionals in banking (57%), insurance (52%), fintech (53%), and wealth management (60%) say the same.

Accountants’ level of concern over the speed at which AI is evolving reflects their perception of the risks of AI use in their industry (see Figure 2). Above all else, accountants fear a loss of personal touch with customers and clients — which, in this customer-focused profession, could undermine their primary value proposition. They also worry the use of AI will ultimately erode essential skills like critical thinking and analysis and introduce new ethical concerns and biases. Of particular concern to the industry are the inaccuracies and misinformation often associated with generative AI: more than four in five respondents (85%) are at least somewhat concerned about generative AI producing nonsensical or inaccurate information.

AI use — and don’t use — cases

Despite their reservations, about half of respondents have at least a moderate interest in AI adoption in their personal or work lives. Of the specific business use cases where accountants would trust AI, the top five are research and fact-checking (55%), assisting employees with routine queries (55%), informational reports for internal use (51%), assisting with marketing and promotional materials (48%), and assisting customers with routine queries (47%). There is much less trust in layoff decisions (2%) and in overseeing and implementing complex mergers and acquisitions (M&A) transactions (4%). In other words, respondents don’t trust AI to actually make business decisions; rather, they expect it to support them on more routine, day-to-day tasks that could benefit from more automation.

It’s important to note that many accountants likely already interact with AI accounting tools and simply don’t realize it. Around one in four (26%) say they are unsure whether their organizational tools have built-in AI. Since predictive AI has existed for decades, there’s a good chance that accountants have been benefiting from it in their technology packages for some time.

“The lack of visibility into the role AI is playing in their current tools may be making accountants leerier than they might otherwise be about adopting new tools, particularly as the technology evolves into generative AI,” says Steve Hasker, CEO of Thomson Reuters. “Generative AI will have a transformational impact on the work professionals do and how it is done but will never replace the human element when advising clients and stakeholders.”

The learning curve

Just as they may be unaware of AI in their current technology, many accountants tend to be unfamiliar with tax-specific AI tools. Many accountants’ personal experience with AI is limited primarily to ChatGPT and Microsoft’s AI offering through Bing. Fewer than 11% have used other common AI tools at home or work.

A primary element in determining use cases and guardrails is gaining knowledge about evolving AI technology. Savvy accounting professionals are using various resources to close this knowledge gap. More than 40% attend virtual events, consume content about AI from general media sources, or consume content from industry-specific media outlets. However, nearly one in five (19%) are not proactively seeking information on advances in AI. These professionals may be at a disadvantage compared to their peers and out of alignment with clients, many of whom expect the kind of capabilities only AI can provide.

Looking ahead

In addition to efficiency and productivity gains, AI has the potential to support talent management, a critical ongoing issue in the industry. For example, professionals can use AI tools to secure accurate, curated, contextualized information, replacing more general search engines or printed reference volumes. Quick and easy access to this information can streamline employee tasks, taking more tedious, less-fulfilling work off their plates and enabling them to focus on higher-value customer relations. All of that can lead to better satisfaction metrics for employees and customers, which in turn tell an enticing story that could appeal to potential employees.

The value of a trusted partner

For accounting firms looking to realize the benefits of AI, partnering with vendors that offer AI-enabled tools with built-in guardrails may be easier than building capabilities in-house. “With talent for in-house implementations challenging to secure, experienced vendors can be a key resource for accounting firms seeking AI solutions that tap into efficiency and productivity gains while minimizing technology risks,” Hasker says.

In the long run, the risk of failing to understand and adopt responsibly implemented AI tools may be greater than the risk of using them. Falling behind competitors and not meeting customer expectations are real possibilities for firms that don’t embrace thoughtful AI technology.


This research was conducted online by Arizent, the parent company of Accounting Today, in January 2024 among 226 accounting professionals across job titles and functions.

About Thomson Reuters

Navigating the complexities of tax law has never been more challenging, placing increasing demands on tax professionals. With time constraints tightening and the risk of oversight looming, the necessity for an intuitive tax research platform has never been more pressing. The complexities that arise from GenAI adoption can be daunting, but that’s not a reason to sit on the sidelines; it’s time to double down and embrace the rise of GenAI. That’s why we have developed professional GenAI to help your organization safely unlock the benefits of adoption with a trusted partner. Recognizing these challenges, we are excited to introduce a new premium offering — Checkpoint Edge with CoCounsel.

Developed in collaboration with research scientists from our Thomson Reuters Center for AI and Cognitive Computing, as well as our expert editors and product designers, Checkpoint Edge with CoCounsel is designed to streamline the tax research process. By studying customer workflows and user behaviors and identifying areas prone to inefficiency and errors, we’ve harnessed the power of AI to create a truly transformative experience.

Checkpoint Edge with CoCounsel uses AI to predict the questions you want to answer, offering suggestions and time-saving shortcuts. These predictive queries not only guide researchers but also serve as invaluable aids in navigating complex tax scenarios. Our innovative platform elevates tax research technology to new heights, delivering pinpoint accuracy and unparalleled efficiency.

Experience the future of tax research with Checkpoint Edge with CoCounsel — where predictive and dialogue-based enhancements coupled with AI-driven analysis redefine the standard for excellence. Transform your workflow and ensure precision with Thomson Reuters Checkpoint Edge with CoCounsel.

About Arizent Research

Arizent delivers actionable insights through full-service research solutions that tap into their first-party data, industry SMEs, and highly engaged communities across banking, payments, mortgage, insurance, municipal finance, accounting, HR and employee benefits, and wealth management. They have leading brands in financial services, including American Banker, The Bond Buyer, Financial Planning, and National Mortgage News, as well as professional services, such as Accounting Today, Employee Benefit News, and Digital Insurance.

For more information, please visit arizent.com.

Checkpoint Edge: AI you can trust

Find the deep tax and accounting insight you need faster with natural-language search and intelligent filtering embedded within your research platform