AI Thought Leaders: Progress, promise and peril

Accounting's AI revolution not only continued into 2024 but actually seemed to accelerate, as it has now become nearly impossible to go to a conference, sit in a strategy meeting or even shop for new software without hearing those two famous letters, often preceded by the word "generative" and followed by the words "powered" or "driven." This might seem rather strange, as around this time last year we were marveling at how far AI has come in such a short time. Yet at the end of 2024 we find ourselves in this place once more as the current generation of AI tools makes last year's seem almost quaint. 

This is why, in our second annual AI Thought Leaders Survey, we asked experts in the field what they thought of the past year. The field of AI is both vast and ever-changing, and we wanted to see what people deeply enmeshed in AI in accounting thought of all the changes they've seen this year. 

Many noted that AI has gone from being a novelty or an experimental tool in many cases to being a practical, widely adopted technology integral to daily operations. In this respect, even those who may not consider themselves tech savvy are now using sophisticated AI tools that would have seemed like science fiction as little as 10 years ago. Strategic decision-making, advanced analytics and personalized client interactions are just the tip of the iceberg when it comes to use cases for accountants. 

"At the beginning of the year, AI in accounting felt like an emerging trend that many were watching from the sidelines," said Kacee Johnson, vice president of strategy and innovation at CPA.com, talking about the noticeable shift. "It's no longer just about automation; the conversations have evolved to exploring how AI can enhance advisory roles, improve decision-making and solve capacity challenges. I've seen more professionals embracing AI as a tool they need to understand and leverage, not just something that might affect their work down the line." 

The speed at which AI has advanced this year impressed many, especially its generative capabilities and its application to data, both structured and unstructured. In a short time it has transformed workflows, increased productivity and uncovered new insights never considered by users. Meanwhile, the recent rollout of specialized AI agents capable of limited autonomy to handle complex tasks like fraud detection, tax analysis and data reconciliation tells them there's still so much more to come. 

"We have all seen AI advance significantly in the past year, especially in the area of automation of manual tasks," said Jim Bourke, managing director of Withum's advisory services. "Think about areas like bill pay, invoicing, expense management, financial statement analysis, etc. AI is putting accountants into more strategic roles and getting them out of the trenches in doing the manual tasks. This past year I have seen a number of players in the tax space surface by leveraging AI. Although many of them still continue to be a work in progress, we are going to see AI totally change the tax space and eliminate the massive tech stacks that exist in many firms today."

Of course, all technologies have their risks, and AI is no exception. Indeed, as the technology's presence in firms grows, so too have the concerns about its use. Experts cited security risks like data breaches and misuse of sensitive information by AI systems, and many are worried about the accuracy of AI output given the tendency to "hallucinate" (i.e., make up stuff). But they also raised broader ethical concerns, such as the perpetuation of bias as well as potential job displacement in the short term. Experts didn't think AI was going to wholesale replace accountants anytime soon, but some conceded it would serve to disrupt job dynamics in certain parts of the profession. 

"It's poised to replace certain jobs or at least automate specific tasks within jobs," said Prashant Ganti, head of product management in Zoho's Finance and Operations business unit. "AI agents will influence particular roles, potentially altering the premium placed on certain skills, leading to some traditional jobs disappearing entirely."  

In this, the first of three parts, we look at what our experts—drawn from CPA firms, software vendors and academics all deeply involved in the field of artificial intelligence in the accounting world—thought of three questions: 

  • How has your perception or impression of AI in accounting changed from the beginning of this year to now?
  • What scares you the most about AI today? What is your biggest concern? 
  • What's impressed you the most about AI this year? What really got your attention (both in terms of accounting and overall)?.

We'll have more from our experts next week.

Davis Bell

CEO, Canopy
Davis Bell
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

Honestly, my perception of AI hasn't changed. At the beginning of the year I felt confident that Amara's Law would play out with AI in accounting, meaning in the short term the impact of AI would be less than expected, and in the long term it would far exceed what we can imagine. I feel the same way now. 

What scares you the most about AI today? What is your biggest concern?

New technology always brings with it both positives and negatives. It's called "disruption" for a reason. I think AI is going to be very, very disruptive — in ways good and bad — and I'm not sure we're ready for that - on both a human level and a societal one. I can imagine AI helping us crack the code on diseases that cause great suffering, but I can also imagine lots and lots of economic displacement, too. I think we have time to get ready, but whether or not we'll use that time is an open question. I hope we do.  

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The speed and scale of improvement in AI is incredible. It's getting so much better, so fast. I believed this would happen, but it's impressive to actually see in practice, and also to see the new capabilities this unlocks for us as a software provider. It's going to allow us to leverage AI to save accountants so much time on menial things, and give them time to use their high-level strategic abilities to help their clients. Imagine regular individuals and small businesses getting service and support and consulting that are currently only available to those with far more money and resources; it's going to be awesome.

Jim Bourke

Managing director, Withum Advisory Services
Bourke-Jim.jpg
Jim Bourke
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the beginning of the year I was very curious about the potential ways that AI could potentially impact our industry. At the beginning of the year I was perplexed about the ways that AI would impact jobs and the role of new hires in our profession. As 2024 rolled on, we learned a lot more about AI as a result of our use of Copilot. When Microsoft removed its 300-seat purchase minimum in early 2024, we rolled out Copilot internally and started to learn that AI would not take away jobs from those in our profession, but simply place them into different roles, as a result of the automation of manual tasks. We also started to see an increased adoption of this technology by our staff and started to see them become vocal on ways that AI could help them in their current roles.   

What scares you the most about AI today? What is your biggest concern? 

What scares me the most about AI is actually my biggest concern and that is around data privacy and security. Since AI is rooted in extracting intelligence out of vast amounts of data in near record time, there is a significant risk of data breaches and misuse of that data, including personal and private information. Today I am good with where we are at, but with the continued rapid expansion of AI, I just hope that the security front can keep up with the development of tools and techniques to keep the bad guys away!

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

We have all see AI advance significantly in the past year, especially in the area of automation of manual tasks. Think about areas like bill pay, invoicing, expense management, financial statement analysis, etc. … AI is putting accountants into more strategic roles and getting them out of the trenches in doing the manual tasks. This past year I have seen a number of players in the tax space surface by leveraging AI. Although many of them still continue to be a work in process, we are going to see AI totally change the tax space and eliminate the massive tech stacks that exist in many firms today.

Samantha Bowling

Managing partner, GWCPA
Bowling-Samantha-Garbelman Winslow
Bonnie Johnson
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

AI has evolved from a tool for optimization to a strategic partner in decision-making. At the beginning of the year, we primarily saw AI as a driver of efficiency; now, it has become a catalyst for innovation, reshaping how we approach client service and internal processes. Each member of our team collaborates daily with their own AI-powered assistant, a personalized bot designed to enhance productivity and creativity. These AI companions elevate our thinking, streamline routine tasks and improve our ability to communicate effectively, allowing us to focus on delivering greater value and deeper insights.

What scares you the most about AI today? What is your biggest concern?

My biggest concern is the potential misuse of AI, particularly regarding data security and decision biases. Without robust ethical guidelines, AI could inadvertently harm trust in the profession.

What's impressed you the most about AI this year? What really got your attention?

AI's ability to simplify complex regulatory changes and elevate client advisory services has been remarkable. Internally, its impact has been equally transformative, driving significant time savings and innovation. We've developed custom ChatGPT solutions to support every facet of our firm—ranging from marketing and tax to software troubleshooting and SOP navigation—enhancing both efficiency and effectiveness across the board.

Ted Callahan

Director of partnerships and strategy, QuickBooks Partners Segment, Intuit
Ted Callahan
How has your perception or impression of AI in accounting changed from the beginning of this year to now? 

As with any new technology, there comes a period of considerations and concerns. However, over the last year, we've seen accounting professionals embrace AI on a large scale. In fact, our data shows that a majority of accountants (98%) have used AI for firm operations in the past year — tapping the tech for everything from managing client portfolios (65%) and client communications (54%) to invoicing and payments (53%). As we look ahead to the new year, AI will continue to give accountants the foundation they need  to provide strategic business advisory services by automating tasks and freeing up their time for higher level work, which is especially critical as the industry continues to grapple with staffing challenges.

What scares you the most about AI today? What is your biggest concern?

When it comes to AI, accountants reported their greatest concerns are data privacy and security (31%), accuracy (21%), and implementation and maintenance costs (21%). On that front, accountants should be cautious when sharing sensitive client information with large language models as this can create risks for exposing sensitive data to the wrong parties. However, even with these concerns, keeping up with emerging technologies like AI is critical for the accounting industry. At Intuit, we're focused on safeguarding data and protecting privacy using industry-leading technology and practices, while adhering to responsible AI principles that guide how the company operates and scales its AI-driven expert platform.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

I'm most impressed by accountants' unprecedented adoption of AI, especially within the broader context that many non-accounting pundits believe it will replace accounting jobs. 

In fact, accountants are largely not concerned with the speculation that AI will replace them. Instead, accountants are leveraging AI capabilities to enhance success for both themselves and their clients, using the technology to streamline workflows and boost overall efficiency. 

AI holds untapped potential, but bringing its potential to life requires creativity and vision. Firms are already tackling this challenge by experimenting with AI to find new ways to bring value to their clients. This shift is especially timely as the industry struggles with waning  talent pipelines, with 94% reporting that hiring has been a challenge in the past year. However, 9 in 10 accountants reported that technology, including AI, could help solve the skills shortage problem by automating tasks and reducing manual workloads for accountants.

Daren Campbell

Tax technology and transformation leader, EY Americas
Daren Campbell
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

Overall, my impression of AI in accounting hasn't changed significantly over the last year, with part of that due to working in the AI and tax space for over seven years now. What has changed is the increase of the actual application. We've moved out of theory and into the hands of how people are using AI, which has shown some of the challenges and hurdles in AI adoption.

While AI will continue to be transformative for our industry, we're at a point now where it's going to be fast in some areas and slow in others. As companies continue to incorporate AI into accounting, the slowness can be identified as being driven by two things: the processes in place and accessibility to data. Companies must change their internal processes to fully take advantage of AI, and right now they have huge data challenges that need to be addressed.

What scares you the most about AI today? What is your biggest concern?

My biggest concern with AI is not the technology itself, but how people use AI, which can be put on two sides. One is an overreliance or trust in AI. When we look at what AI does and does well, we think that because it's good at certain tasks and activities, it's good at everything. This leads to overreliance or over-trust without an understanding or critical questioning of the results that are coming out of the AI.

On the other side of it are the risks associated with bad actors that will use AI to accelerate their motives and purposes. In and of itself, AI is not bad, but in the hands of bad actors, it can create a lot of risks, especially with the acceleration of the technology around programs like deep fakes and digital twins. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

In this last year, the speed at which AI has advanced has been most impressive. Especially when seeing all the major actors involved in AI and the significance of each advancement, including large language models and GenAI tools, the quickness of these advancements has caught my attention, as well as how quickly large challenges within have been addressed. 

When we look at pretrained models and some of the challenges with subject matter specialists, there have been a number of techniques that have evolved to develop and train small language models to use things like retrieval augmentation generation, which has the ability to actually help AI be smarter in a focus area. Overall, the speed of a lot of these improvements has been quite impressive, and that's true in both accounting and overall.

Jack Castonguay

Vice president of strategic content development, Surgent
Castonguay-Jack-Knowfully Learning Group
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

When it first was introduced, I thought AI was largely Apple's predictive text on steroids. I wasn't impressed. Now, I see its benefits and am excited for what comes next. Its ability to analyze large datasets, identify potential frauds, and its ability to easily "think" like a person make its use case in accounting almost endless. What used to take half a day, like summarizing a new ASU, now can take mere minutes even with a quality review. I think we are only beginning to scratch the surface of how AI in accounting can make our lives easier and our work products more accurate. I'm impressed now and only expect that impression to grow.

What scares you the most about AI today? What is your biggest concern?

AI in its current state does not scare me because it is just a powerful tool. However, potential future versions scare me because of what they may mean for knowledge work. We have long told students and staff that if you know more than others and if you can be a trusted resource for your client, then you will be successful. We have placed a high value on knowledge, critical thinking and intelligence in the workplace. But what happens when AI can easily replicate that knowledge and intelligence? It devalues the humans who possess it. As a society we are set up to value and reward the thinkers: from Einstein to Oppenheimer to Steve Jobs. We are not prepared to operate in a society where those things or people can be replaced by AI. That keeps me up at night.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

What impressed me most was the embrace of AI by the accounting industry. We usually aren't known for being early adopters or technology leaders, but in the AI space, many firms and practitioners jumped in headfirst and adopted AI early and used it often. Ex-ante I wouldn't have expected the excitement and use around it. AI's use in war by some adversaries to select, identify and strike targets got my attention for its negative uses more than anything else. The same tools that are great for improving our coding in Python or summarizing the finer points of a new AICPA standard are devastating for those in war zones.

Danielle Supkis Cheek

Head of analytics and AI, Caseware
Cheek-Danielle-PKFTexas.jpg
How has your perception or impression of AI in accounting changed from the beginning of this year to now? 

There have been significant advancements in transparency and maturity in AI, making it much less a "shiny object" and more relevant to daily users across audit and accounting. The ability to understand data provenance and ensure secure, structured applications is bridging the gap between innovation and practicality. AI is now actively transforming our approach to work — not just as a flashy new tool — but as a meaningful way to solve problems more efficiently.  

What scares you the most about AI today? What is your biggest concern?  

While many of the recent innovations in AI are actually focused on making things safer and more transparent for users, the sheer speed of innovation could be overwhelming. Even for those embedded in the tech space, keeping up is daunting, and for end-users, it's even harder. This is why balancing innovation with safety is critical. It's great that AI can disrupt roadmaps and push us to rethink long-term plans to solve problems differently, but we've got to stay alert and agile to maintain a healthy balance between innovation and user protection.   

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.  

The way generative AI is handling structured and unstructured data is helping us build entirely new approaches to solving problems that are completely different from the way we've done it historically, and which would have previously taken years to build out. We can now restructure semi-structured or structured data in ways we couldn't before. My entire view of what data is and what can be done with it with Gen AI has expanded dramatically over the course of the year. It's amazing what you can do with a well-crafted prompt. 

Ellen Choi

Founder and CEO, Edgefield Group
Ellen Choi
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the start of 2024, AI was an exciting buzzword, met with both curiosity and skepticism about its relevance to accounting practices. Since then, the landscape has shifted—AI is now universally recognized as critical, even if many remain uncertain about how to implement it effectively.

What's emerging is a Tale of Two Cities: a growing divide between the "haves" and "have-nots." While the Big Four and enterprise CPA firms leverage dedicated AI teams and resources to drive measurable gains, smaller firms often lack the knowledge, resources, or clear strategy to keep up.

Innovative vendors like FieldGuide, Rightworks, Aiwyn and Material.ai (acquired by Thomson Reuters) are advancing the field with generative AI and AI agents, but adoption remains uneven. These breakthroughs haven't yet trickled down across the profession.

The potential for AI in accounting is vast, but ensuring its benefits reach across the entirety of the profession is the next significant challenge.

What scares you the most about AI today? What is your biggest concern?

What concerns me most is that firms won't move fast enough to pragmatically apply and embrace the benefits of AI. While it's true that AI is a powerful tool that must be used thoughtfully and with caution, caution shouldn't become an excuse to delay action. The longer firms drag their feet, the wider the gap grows between those adopting AI and those left behind.

Younger workers are already using AI tools, whether firms encourage it or not. They expect workplaces that embrace modern technology to enhance their productivity and skills. If firms don't catch up, these employees will seek opportunities elsewhere, gravitating toward organizations that empower them with cutting-edge tools. This isn't just about staying competitive with clients—it's about staying competitive in attracting and retaining the next generation of talent. The future workforce demands progress, and firms must rise to meet that expectation.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

This year, I've been most impressed by the rapid advancements in AI agents and multimodal applications. The pace of progress among leading AI developers like OpenAI, Microsoft and Google has been both astonishing and terrifying, reshaping the landscape of AI capabilities at breakneck speed. Technology startups, including those in accounting, are already deploying and training AI agents for specific tasks in production.

What's equally exciting is the emergence of multimodal applications that go beyond text. This is truly a next-level value unlock for accounting firms. Technologies like AI voice generation via NotebookLM and Eleven Labs, and generative-AI enabled video/digital clones via Heygen and Synthesia are no longer theoretical—they're in the wild.

The challenge now is identifying the right use cases, applying these tools thoughtfully, and ensuring guardrails are in place to guide their adoption responsibly. The possibilities are vast, and the firms that act strategically and boldly stand to gain a significant competitive advantage.

Sergio de la Fe

Enterprise digital leader, partner, RSM US LLP
Delafe-Sergio-RSM.jpeg
Sergio de la Fe
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the beginning of the year, I was uncertain about how AI would benefit the accounting profession. Over the past 18 months, many organizations have been focused on becoming informed, testing new solutions, running structured proof of concepts, and ensuring the necessary support and governance to scale AI safely, where it fits and creates value.

As the year progressed, it became clear that AI can significantly enhance both talent and client experiences by improving efficiency, effectiveness, and quality. I now firmly believe that AI will have a transformative impact on the profession and the way firms serve and add value for their clients.

Looking ahead, there is optimism about AI's potential to drive innovation, reshape the accounting landscape, and the need to continuously develop the competencies required to advance alongside this technology.

What scares you the most about AI today? What is your biggest concern? 

There isn't anything about AI that is inherently frightening. Rather than concerns, I see AI as full of opportunities. As with any new technology, especially one as disruptive as AI, there are evolving situations to consider when leveraging and managing it.

The strategy involves assessing risks and mitigating them where possible. Educating people and clients about the current state of AI, the tools being used or planned for use, and empowering teams to work with AI rather than fear it is key. By fostering this understanding, challenges can be proactively addressed, enabling teams to embrace AI as a valuable resource.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

What's most impressive about AI this year is its potential to increase quality and enhance capabilities in accounting and broader business operations. In accounting, AI's ability to assist with routine tasks, improve accuracy, and uncover insights from data has been transformative. The accessibility of AI for the middle market is especially notable, enabling businesses to leverage advanced technology to drive efficiency and growth. Across industries, AI's potential to foster innovation, enhance decision-making, and create new opportunities is truly exciting. It's clear that AI is no longer a distant future—it's here and making a real impact today.

Avani Desai

CEO, Schellman
Desai-Avani-Schellman
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

Honestly, it hasn't changed much, though I have become more bullish. Having seen so many companies leaning into AI to overlap it with their existing capabilities, their productivity gains are real—and so are ours. As the CEO of a tech-enabled company that specializes in tech assessments, this is exactly the future I was hoping for, and it's rewarding to see it happening in real time.
 
What scares you the most about AI today? What is your biggest concern?

My biggest concern is data privacy and security. AI systems require massive amounts of data to train, and that opens the door for potential misuse or breaches. To prevent that, we need to start answering critical questions like, who owns the data? How is it being used? How secure is it? I'm also concerned about the ethical implications—bias in AI models and how decisions are made without human oversight. If we don't address all of these issues thoughtfully, the consequences could be significant.
 
What's impressed you the most about AI this year? What really got your attention?

It's incredible how quickly people have embraced AI. I remember when the cloud was first introduced—it took years, even decades, for people to fully trust and adopt it. But AI? Wow. Its value is so apparent that adoption is moving at lightning speed, especially in industries like accounting, auditing, and assessments. That kind of momentum is amazing—it really underscores how transformative the technology is.

Pascal Finette

Founder and CEO, Be Radical
Pascal Finette
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

I believe that whereas AI, at the beginning of 2024, was largely still a novelty and something some of us "played with," it has become much more a thing we now "do." Concretely, I vividly remember running an AI workshop during my DCPA session in late 2023, where the vast majority of participants had, at best, toyed a little bit with ChatGPT, and only a handful of people paid for the pro version. Fast forward to today, and I rarely meet someone who is not using ChatGPT (or any of its many brethren) in some capacity on a daily basis. I think 2024 can be summarized, when it comes to AI, as the year AI became a tool instead of a curiosity.

What scares you the most about AI today? What is your biggest concern?

I do think that we are going to see some pretty strong shifts in employment in certain sectors — just yesterday, I came across a lengthy discussion on Reddit about a radio show production team that was fired and replaced, unfortunately, by AI. The problem is not necessarily the fact that these jobs are changing (and in some cases disappearing) but the speed with which this is happening and the fact that, societally as well as politically, we are woefully unprepared for this. Most of the discussion around this topic is either politicized, end-of-the-world talk or, on the other side, hand-waving "AI will take care of it" blabber. What is needed, in my eyes, is robust, constructive dialogue, followed by swift action.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

AI-powered search tools (or more precisely, answer tools) such as Perplexity AI have become incredibly good and useful—both for everyday questions such as "What is the best vacuum cleaner?" or "What are the best things to do when in Rome for two days?" but also for researching and referencing tax and accounting questions. Google's research assistant NotebookLM has become an indispensable tool for processing large amounts of information in novel ways (their "turn a PDF into a podcast" feature is incredibly useful to quickly get the gist of complex whitepapers). And Anthropic's first version of Computer Use (where Claude can control my computer) points to an exciting future of easily accessible and highly customized process automation.

Prashant Ganti

Vice resident of global product strategy, development and alliances — Enterprise Finance Suite, Zoho
Prashant Ganti
How has your perception or impression of AI in accounting changed from the beginning of this year to now? 

Largely, we've seen instances of AI being reliably used for automating and simplifying routine tasks like data entry, collation, and generation of simple reports. But, over time, it is becoming more evident that AI can offer more than just automation of repetitive tasks. AI is actually proving to be a valuable strategic assistant that provides predictive analytics, detects fraud, and identifies risks, uncovering deep insights and patterns that would otherwise be cumbersome to find on our own. For example, AI can be used by organizations to identify anomalies and highlight potential areas that could lead to financial and compliance risk, offering auditors a level of insight that would typically require intensive manual effort to uncover. 

What scares you the most about AI today? What is your biggest concern? 

I think we all need to be reminded that AI is a technology, which means that it can be prone to errors. Granted, the system becomes more sophisticated with increased usage, but it can never be infallible. My biggest concern is that we start losing accountability and become blind to its biases and harmful patterns since the technology is starting to permeate many high stakes areas like finance and healthcare. We need to use AI more conscientiously. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

I think the rise of AI-powered tax advisory tools is most impressive. It serves as an assistant to accountants and tax experts, helping them answer the most complex client questionnaires on taxes, create memos, do audit checks and communicate with clients. These AI tools can go beyond all this to also glean financial data, predict future tax liabilities, and provide tailored recommendations for maximizing deductions. 

Since the only thing constant about tax is change, these capabilities ensure the clients are always up-to-date with their strategies and compliance. More importantly, tax experts and accountants can focus on providing high-level advisory services while leveraging AI to reduce the amount of manual effort involved in many of these important tasks.

Mike Gerhard

Chief data and AI officer, BDO USA
Mike Gerhard
How has your perception or impression of AI in accounting changed from the beginning of this year to now?  

At the start of the year, AI in accounting felt like an exciting frontier, full of potential but still largely uncharted. Fast forward to now, AI has evolved from a realm of exploration to a focus on intentional solutions. Its integration into routine tasks is proving valuable, enabling professionals to concentrate on more strategic and rewarding work. Furthermore, AI is poised to change the work experience, unlocking the full potential of our professionals and delivering outstanding value to our clients. What's truly exciting is the cultural shift. People are embracing AI, seeing it as an ally rather than something to fear. This acceptance is evident in our own AI solutions, where monthly usage has nearly doubled compared to this time last year. It's a thrilling time to be at the forefront of AI in accounting, witnessing firsthand how it's reshaping the landscape and enhancing the work experience for everyone involved.

What scares you the most about AI today? What is your biggest concern? 

What concerns me most about AI today is the rapid pace of technological advancement. This swift development necessitates critical considerations, particularly in ensuring that our systems and policies for data privacy, security, and bias mitigation in decision-making remain up to date. Responsible AI development and deployment is a top priority. We need robust safeguards to prevent unintended consequences and maintain the trust of our people, clients, the public, and the capital markets. As we advance our AI strategy, we are committed to thoroughly assessing all potential risks. It's vital to keep people at the core of AI decision-making, ensuring these technologies meet the needs of their users. At BDO, we're tackling this with our AI Governance Committee, which is built on the key principles of BDO's Responsible AI Policy, which is our global approach to the responsible management of AI systems to support business growth and quality, while managing the unique risks associated with AI use. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

What has impressed me about AI this year is the overwhelming demand for its integration into our work processes. It's not surprising, but it's certainly impressive to see everyone in our organization, from the CEO to our interns, recognizing the transformative power of this technology. Our teams are diligently working to deliver custom solutions using AI in ways we haven't explored before. In accounting, for example, we've successfully harnessed AI to tackle challenges like boosting employee productivity and revenue growth enablement through Generative AI. The rapid development and deployment of AI solutions have allowed us to address business issues that were previously challenging to solve. This isn't just evident in our own solutions; we're also seeing it with our customers and vendors. These innovations highlight AI's potential to revolutionize our industry and underscore its game-changing nature across the board. It's an exciting time to witness how AI is reshaping our approach to problem-solving and driving unprecedented growth and efficiency.

Chris Griffin

Managing partner of transformation and technology, Deloitte & Touche LLP
Chris Griffin Deloitte
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

AI continued to evolve rapidly this past year. The Large Language Models behind GenAI are more capable of understanding context, generating human-like text, and performing complex tasks. GenAI continues to transform the accounting and audit profession, driving automation, enhancing accuracy, and enabling more strategic decision-making. These advancements are helping audit and accounting professionals to focus on more complex tasks, while reducing the time spent on routine and repetitive tasks.

What scares you the most about AI today? What is your biggest concern?

GenAI is still in its infancy, and it carries risk. Some of the most important risks to address relate to privacy and security, managing bias, transparency and traceability of results, and IP ownership. As with any innovative technology, malicious actors can abuse GenAI for forgery, phishing and fraud. GenAI's large language models synthesize vast amounts of data so quickly that it can be challenging to track its sources, making it difficult for humans to understand the validity of the information it produces. 

Human oversight to manage privacy, security, legal and behavioral risks is critical to capitalizing on GenAI's remarkable capabilities. For instance, there is no single process in an audit that can be automated end to end because of the judgment required at conclusion. The ultimate review and conclusion of GenAI-powered tasks requires a skilled, experienced professional to reach a professional judgement. At the end of the day, an audit opinion bears the name of a professional, not a machine. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

AI's capabilities continue to evolve—from standalone GenAI applications to AI agents that can reason, plan, remember, and act autonomously. Unlike typical language models that automate individual tasks, these AI agents can automate entire workflows and processes. This represents a shift in AI being deployed as a supportive tool to being leveraged as one that is more collaborative. While AI agents can unlock enormous potential, robust governance frameworks are critical to manage risks, including data ethics, privacy, and security.

Aaron Harris

Chief technology officer, Sage
Aaron Harris
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

I've been very impressed with the industry's response to AI.  Rather than viewing AI as a threat, industry and practice leaders are embracing AI. In our research, 43% of firms say they expect to hire more staff as a direct result of the new opportunities AI presents.

What scares you the most about AI today? What is your biggest concern?

Despite the awesome potential, AI is an imperfect technology.  The problem is this particular wave of technology disruption has almost no barriers to adoption, so there's a massive proliferation of tools and solutions, not all built with appropriate care and safety.  Bad experiences are inevitable and it will only take a few for users to write off the tech.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The emergence of AI agents — AI that is capable of taking on complex tasks; requiring reasoning and specialized skills like coding.  These agents will liberate humans from tedious tasks, bringing humans into the loop as needed to make decisions and review exceptions. 

Wesley Hartman

Founder and CEO, Automata Practice Development
Wes Hartman 2
Sonia Alvarado
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

My impression of AI has not changed much. AI is getting better and better, but it is improvements on the existing concepts. However, I think we are starting to see the implementation of AI models in the accounting space into new apps that are built around AI. At the beginning of the year, many existing apps in the space were adding AI but it was a simple implementation. As all these existing companies were adding to their websites that they were power by AI, I was skeptical if they actually were powered by AI or just riding the marketing hype. As we start seeing start ups come out of stealth mode, we will see apps that were built with AI first.

What scares you the most about AI today? What is your biggest concern?

My biggest concern on AI is the risks and costs that are still not talked about enough. Not just data risks around whether or not to put client data in but more around the energy consumption and human cost. The energy cost is greater than some small nations and continues to grow. While some AI models are working on training themselves, most including the big AI companies still use humans to review and approve the content scrapped from the internet including all the negative parts of the internet that many of us don't go to. My hope is that we work on new sources of energy and better ways to train the models that don't include some of the darker parts of the internet.

 What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The number of startups jumping into the accounting space has been great. I feel like new tech for accounting has always been an afterthought for Silicon Valley. I think AI has equalized some of the barriers to entry. We will see what shakes out of the race but it is exciting that there is a renewed interest.

Joel Hughes

CEO, Rightworks.
Joel Hughes
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

The initial buzz around generative AI seems to be settling into a more realistic assessment of its potential within the accounting profession. While many tech companies are developing AI solutions, we haven't yet seen the disruption that was initially anticipated. And yet, accountants are judiciously taking an interest and actively experimenting with it.

The profession is also recognizing that meaningful AI integration requires technological advancement, changes in processes, skills and regulatory frameworks. There's a growing focus on developing AI solutions that address specific pain points rather than attempting to revolutionize the entire field. We believe the pragmatic examples promoted by industry thought leaders are showing firms the "doable" possibilities which makes it easier for accountants to get onboard. Just as we were taught to be effective with spreadsheets and PC-based accounting applications, generative AI will become another standard tool we use where it makes sense for us individually. 

What scares you the most about AI today? What is your biggest concern?

One of my primary concerns about AI is the potential for confusion and misinformation in the marketplace. The misunderstanding of AI's actual capabilities and limitations, coupled with outdated perceptions, can prevent experimentation and implementation. Additionally, generative AI tools do a remarkable job of "hallucinating" or providing convincing answers that seem factual. This can be a significant issue where the stakes are high for mathematical or advisory errors. I believe it's important to use generative AI tools that have been "purpose built" (i.e. specific tax, advisory, or assurance GPTs) and on which the accountant is also trained to use. Firms would do well to develop AI usage procedures and guidelines, similar to how audits and tax returns are competently produced. 
 
What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

We've seen impressive AI advancements across various fields in 2024. Companies that have built the most capable foundational models are leading the way in innovative applications of GenAI capabilities with more advanced developments on the horizon. For example, Claude's Artifacts feature enables non-technical users to create mini web-apps for data visualization and concept testing. OpenAI's o1-preview model boasts improved reasoning capabilities, hinting at more sophisticated AI problem-solving.  

In the accounting profession, we've seen interesting AI product announcements, but the industry-disrupting breakthrough hasn't happened yet. However, substantial investments indicate that major transformations in accounting practices are on the horizon. It's only a matter of time before we witness truly game-changing AI applications in the field.

Kacee Johnson

VP of strategy and innovation, CPA.com (an AICPA company)
Johnson-Kacee-CPAcom NEW 2022
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the beginning of the year, AI in accounting felt like an emerging trend that many were watching from the sidelines. Now, there's been a noticeable shift—accounting and finance professionals are diving deeper, with greater awareness, understanding and curiosity about its potential. It's no longer just about automation; the conversations have evolved to exploring how AI can enhance advisory roles, improve decision-making, and solve capacity challenges. I've seen more professionals embracing AI as a tool they need to understand and leverage, not just something that might affect their work down the line.

What scares you the most about AI today? What is your biggest concern?

What scares me most about AI today is how bad actors could exploit its power. From generating convincing deepfakes to orchestrating sophisticated phishing attacks or manipulating financial systems, the misuse of AI poses a serious threat. Its ability to scale and automate harmful actions faster than we can respond creates vulnerabilities across industries, including accounting and finance.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

What's impressed me most about AI this year is the sheer speed of its maturity. In October 2023, we built a tax research use case by uploading a large document into Claude—a significant breakthrough at the time. Now, just a short time later, the market is flooded with specialized tax research tools, and many firms have even developed their own LLMs tailored to this practice area. The rapid evolution from experimental applications to robust, industry-specific solutions has been remarkable, both within accounting and across other industries. It's a clear testament to how quickly AI is reshaping the landscape.

Jenn Kosar

US AI assurance leader, PwC
Jenn Kosar
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

We're excited by the enhancements to productivity we have already seen throughout the year and the enthusiasm with which our people have embraced the possibilities of AI. From our conversations with clients, particularly in finance functions, it is no different throughout the profession. It has sparked even more innovative ideas on how this technology can enable more transformative solutions as the technology matures and becomes fully integrated not just in finance but throughout organizations.

What scares you the most about AI today? What is your biggest concern?

AI systems are becoming increasingly complex, consuming significant amounts of data to generate outcomes that can, at times, be unpredictable or difficult to understand without human knowledge and expertise, and at times even with data science expertise. Despite the promise of GenAI's accessibility to all, some organizations are not necessarily keeping pace with the requirements for governance, risk management and modified controls as they move forward with AI strategies.  

Our approach to Responsible AI is human-led and tech-powered. Accountants and auditors bring the knowledge and expertise to effectively increase opportunity and value of AI while addressing potential risks. By training accountants and auditors to adopt Responsible AI principles and implementing them throughout the business strategy, companies can build a foundation that educates employees and stakeholders on the capabilities of this technology.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

AI is spurring the development of new classes of products and services. At the same time, AI is transforming the product development process. Companies are using AI to speed product design and development with powerful brainstorming, iterating and solutions, similar to AI's transformation of software development and other enterprise functions. The technology is improving rapidly, and our ability to develop guardrails and solutions to improve its performance is rapid as well — making it even more useful in a business context even faster than we thought.

Thomas Mackenzie

KPMG US and global chief technology officer, KPMG
Thomas Mackenzie
How has your perception or impression of AI in accounting changed from the beginning of this year to now?  

With how these technologies are driving innovation, I really believe we are in an "AI era" of financial reporting, where the use of AI and GenAI is becoming ubiquitous as firms invest in and deploy new capabilities.  

According to our latest report on AI in finance, the use of AI in the finance function has reached its highest levels yet, with 62% of US companies using AI to a moderate or large degree, 58% piloting or deploying GenAI and 52% using AI specifically in financial reporting.
 
Over the past six months, finance functions have made the most progress using AI for research and data analysis (60%), predictive analysis and planning (55%) and fraud detection and prevention (54%).  

Looking ahead, I think the benefits of AI will only accelerate — in the next three years 53% of finance leaders expect AI to enable faster access to relevant information and data.   

What scares you the most about AI today? What is your biggest concern?  

While companies anticipate AI will enable better, faster insights, over half (57%) report data security and privacy are the biggest AI adoption barriers. 

These technology-centric risks (cloud, cyber, AI, etc.) pose the biggest potential threats to the integrity of financial reporting, and proper AI governance is necessary to ensure we leverage these capabilities responsibly.  

The good thing is, a majority of companies (66%) are developing corporate principles and guidelines on responsible AI use, and 62% want their external auditor to conduct risk mitigation and internal controls on their behalf using AI. 39% plan to include AI risk and associated controls in the scope of their financial reporting processes over the next year. 

At KPMG, our Trusted AI approach ensures we deploy AI with a human-in-the-loop mindset to maintain quality and accuracy. Our teams are trained to use AI as an enabler while critically engaging outputs and exercising professional skepticism. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.  

What really stood out to me was how the benefits of AI compound the more firms leverage it. AI is playing a major role in enhancing quality, driving innovation and uncovering deeper insights.  

As finance leaders see returns on their AI investments, they are more inclined to shift spending priorities towards an AI-enabled financial reporting process. 92% of U.S. companies report their finance function's AI initiatives are meeting or exceeding ROI expectations, and more than one-third plan to increase AI budgets or shift funds from other activities to drive further AI adoption.   

At KPMG, AI is freeing up resources for our auditors to spend more time on the areas of highest risk, sector-specific risks and challenges. We've embedded our GenAI assistant directly into KPMG Clara, our global smart audit platform, and like an additional team member, it helps auditors refine risk assessments, develop substantive testing procedures and enhance audit documentation.

Blake Oliver

CEO, Earmark
Oliver-Blake-2024.jpg
Blake Oliver
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the beginning of the year, I believed in the potential for AI to automate most of the work accountants do today. During the year, I experienced it firsthand. We successfully implemented a workflow at my organization that reduced the cost of a process we run 100 times per month by a factor of 100. In other words, a process that would cost us $20,000 per month to do with humans now costs $200. It validated what I already believed was possible. I'm confident that AI will eventually automate 80% to 90% of accountants' work within the next 10 to 20 years.

What scares you the most about AI today? What is your biggest concern?

The potential for abuse worries me the most. AI is going to make fraud a lot easier, but it's also going to make detecting and preventing fraud easier. There will be an AI technology arms race when it comes to financial crime.

What's impressed you the most about AI this year? What really got your attention?

What's impressed me most is the extremely rapid improvement in the LLMs from ChatGPT and Anthropic. The difference between ChatGPT 3.5 to Chat GPT 4o and Claude 2 to Claude 3.5 Sonnet is almost unbelievable.

Adam Orentlicher

Chief technology officer, Wolters Kluwer
AdamOrentlicher.jfif
How has your perception or impression of AI in accounting changed from the beginning of this year to now? 

The accounting industry has evolved dramatically in 2024, with AI moving beyond being just a "strategic partner" to becoming a critical component of practice. While initial concerns about job displacement remain, we're seeing a shift toward "AI-enabled professionals" rather than "AI versus professionals."  71% of firms now recognize AI as a driver of increased profitability. The focus is moving from basic automation to leveraging AI for optimizing the firm, elevating professionals, and enhancing client service.  

Source for 71%: https://www.paymentsjournal.com/accountants-view-ai-as-an-ally-not-a-competitor/

What scares you the most about AI today? What is your biggest concern? 

Two primary concerns dominate:

Data privacy and security: Beyond traditional security concerns, we now face challenges with Large Language Models and their potential to retain or reconstruct sensitive information. This demands sophisticated 'walled gardens' and zero-trust security architectures to ensure client confidentiality.

AI reliability with professional judgment: While AI capabilities have grown exponentially, particularly in pattern recognition and predictive analytics, the risk of AI hallucinations in tax & audit contexts requires rigorous verification protocols. This profession demands accuracy.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

The integration of AI in tax and audit is transformative. Systems can now simultaneously process data, documents, images and natural language prompts to provide comprehensive insights. We're seeing significant efficiency gains in routine tasks while improving accuracy in complex compliance scenarios. Most impressive is AI's ability to surface insights from a myriad of data sources that experienced advisors, auditors, partners, staff might miss.

Abigail Parker-Zhang

Accounting professor, University of Texas at San Antonio
Abigal Zhang
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

AI is expanding its role in accounting. Typical use cases of AI in accounting include using machine learning to perform predictions like forecasting revenue or fraud. Now, Gen AI is offering new applications. For example, big accounting firms announced that they have been using Gen AI to do intelligent search, summarize documents, and draft documents.

What scares you the most about AI today? What is your biggest concern?

AI can generate misinformation or inaccurate information and they can be hard to identify. For example, my colleagues and I asked ChatGPT the same question and got different answers. We ended up looking up academic research to verify the answer. This worries me because while developing critical thinking to evaluate AI's responses is often suggested, this is easier said than done. In practice, many users may not notice inaccuracies or take the extra step to verify information. Although AI often provides accurate answers, the uncertainty of when inaccuracies occur makes over-reliance on AI a genuine concern.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

I am particularly impressed by the recent advancements in AI's multimodal capabilities. Tools developed by OpenAI and Google Gemini demonstrate that AI can now interact with humans not only through text but also via voice, images and video. This represents a significant leap forward, bringing AI closer to the intelligent agents often depicted in science fiction. In accounting practice, this multimodal capability holds great potential. For instance, an AI system that processes data from images, voices and videos simultaneously could greatly enhance risk assessment in audits.

Hitendra Patil

Founder and CEO, Accountaneur
Hitendra-Patil-AccountantsWorld
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

Coming into 2024, I thought of AI in accounting as mostly an efficiency/productivity booster, a faster research assistant and an ideation tool. I was somewhat skeptical about the deeper integration of AI into accountants' day-to-day work. 

Now, my perception has somewhat changed. AI is an absolute must-have partner at accounting practices for automating mundane tasks and elevating the game through advanced analytics and strategic insights. It is not for substituting what the accountant does but for augmenting the accountant's impact. AI now helps ensure better compliance. AI can minimize the grey areas with real-time checks. The focus is shifting from learning about AI to leveraging it. Tech-savviness has become as crucial as number-crunching. AI's capability for personalized financial advice is enhancing accountants' interactions with their clients. I think AI is fast going from a side tool to a central part of accounting practices. 

What scares you the most about AI today? What is your biggest concern?

The privacy of client data. AI systems are trained on underlying data. What if they compromise confidentiality? That will be a trust breaker in the accounting profession. 

AI's reliability and lack of explainability in number-crunching, predictions/suggestions /insights /advice is a worry too. What if it hallucinates without apparent symptoms? When complex accounting rules are in play, or if AI systems are fed biased or synthetic data, the analytics can be fudgy. 

But what scares me the most is AI sidelining human judgment. 

It's not all about accounting data and insights; it's also about ethics, questioning, and viewing the larger picture by connecting the dots between the data world and the one outside it — i.e., the human minds. That human touch, that intuition, is something I think AI won't grab for at least the foreseeable future. We must ensure AI "assists" and not make us lazy to claim that AI-generated insights are our professional creations. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

Most definitely, the way AI is getting absorbed into the software solutions accountants commonly use. 

Such solutions have in-built AI trained on the data that accountants have been creating for years. The ability of AI to preempt the questions that users would/should ask is fascinating. That's game-changing. It's like accounting intelligence on steroids. The potential of catching compliance issues while processes are still not completed is phenomenal.

But what got me was how AI is helping us make sense of everyday chaos. For example, AI pulls insights from a jumble of emails or snippets of conversations. It is like real-time analysis and auditing of human interactions.

Enzo Santilli

Principal, Grant Thornton Advisors LLC
Enzo Santilli
How has your perception or impression of AI in accounting changed from the beginning of this year to now?  

At the start of 2024, there was a lot of focus on the need to build new, proprietary solutions — and those who had first-mover advantage were positioned to win. As the year evolved, it became clear that the more likely route is going to be organizations using AI tools to more effectively mine, integrate and summarize data they possess today and to improve current capabilities, rather than build anew. Also, 12 months ago, AI was thought to be fine for prosaic content summarization but bad at math. That's no longer true.

What scares you the most about AI today? What is your biggest concern? 

Any technology can be used for good or for bad, and AI is no different. What makes AI more challenging is the speed and extent at which it can operate, coupled with a lack of easy auditability. In modern economies, it will by far be a better force for good. But as cybercrimes give rise to the need for advanced information security, there will need to be controls and policing mechanisms for those that choose to use AI for harm.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

The use of AI in scientific research and the search for drugs or cures for debilitating diseases has been interesting to watch. We've only begun to scratch the surface as those models are released. The advances in fraud control and continuous monitoring were the ones most affecting the practice of accounting, and companies have started adopting AI in their own financial planning and analysis functions to perform scenario modeling and variance analysis.

Doug Schrock

Managing principal of artificial intelligence, Crowe LLP,
Doug Schrock
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

In 2024, there have been really solid advances in the functionality of AI tools available to accounting professionals. For example, the year started with wholly unreliable results when applying LLMs to Excel and performing math. But now, LLMs can generate Python code to perform calculations, and user tools such as Microsoft CoPilot are delivering increasingly functional applications such as Microsoft CoPilot for Finance.  AI is emerging right now as critical to accounting as Excel was in 1985.

What scares you the most about AI today? What is your biggest concern?

My concern is that most companies have not yet internalized the level of disruption that AI will bring to their business with the introduction of new competitors, changed buyer expectations, and new efficiency pressures. The differentiation between AI leaders and laggards will start slowly than quickly accelerate leaving many companies and individuals exposed. There are going to be both big winners and losers and I'm concerned leaders are under-indexing in building AI disruption into their strategy. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The start of 2024 was about task-specific, generative AI. The end of 2024 is about autonomous, reasoning, Agentic AI. This is going to be the focus of 2025 and where we will start to see not just tasks being covered by AI, but much more sophisticated, multi-step knowledge worker tasks performed by AI.

Eitan Sharon

SVP of data and science, Xero
Eitan Sharon
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

That we're only scratching the surface of the value that AI can bring to accounting. Our conviction is that AI has the potential to reimagine accounting, and that belief has been grown over the past 12 months. 

More broadly, what has shifted is the focus on the value AI brings, rather than the latest and greatest LLMs. Despite the progress being made in AI, many accounting firms and small businesses are still not fully aware of its potential and how it can add value. 

Over the past 12 months, accounting firms have started to look at more practical ways to plug AI into accounting workflows such as accounts payable and client service. I'm really excited about this marked shift towards genuinely helpful innovation.

What scares you the most about AI today? What is your biggest concern?

While new developments in technology can be unsettling, I don't think fear is the right posture to take toward AI. If anything, I'm optimistic about the opportunities for AI to improve the lives of small businesses and the advisors who help them thrive. We're working hard to bring the benefits of AI to Xero's customers, and that's incredibly motivating for me.

As a field, trust is an area we must maintain a strong focus on. While AI is incredibly powerful, it can still make mistakes without the right guardrails and systems around it. Building trustworthy AI is a key ingredient in delivering value to customers, who need to trust that these tools won't make critical mistakes that hinder rather than help them. 

That's why we've invested in guardrailing products like our small business companion Just Ask Xero (JAX) through proprietary technology called JAX Assure, so our customers can trust the accuracy of its work.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The innovative ways in which accountants are putting AI to work. They're using AI-powered chatbots to chat with their accounting ledgers to track account balances and financial transactions. They're also using AI-driven customer portals to quickly surface information based on user queries and provide summaries of knowledge-based content to help customers more efficiently.

AI is also helping accountants provide higher-value services in areas such as strategic advice, forecasting and financial planning. Tools such as Syft Analytics, an AI-first reporting platform which Xero acquired earlier this year, equips businesses with in-depth insights including custom reporting, visualizations, benchmarking, live analysis views with drill down, budgeting and forecasting and data quality assurance. It empowers small businesses to communicate and advise their clients strategically, eliminating the need for extensive data analysis teams. 

We've seen how AI can be a game changer for firms in responding to client needs quickly and in adapting to a changing market.

Donny Shimamoto

Founder and managing director, IntrapriseTechKnowledgies
Donny Shimamoto of Intraprise Techknowlogies
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the start of the year, I felt like a lot of Gen AI was still a lot of hype. But within this past year, I've seen many vendors incorporate it into their products and provide examples of use cases that really make sense and can help reduce the workload for accountants. I also love the progress that Microsoft has made with Copilot and its integration with the Microsoft Office suite. It's made access to Gen AI something that we use on a daily basis.

What scares you the most about AI today? What is your biggest concern?

With all the AI vendors rushing to pass each other, I really worry that first-to-market profit will be made a priority over privacy and security. This means that CFOs and auditors need to be even more vigilant about ensure that ethical concerns related to the use of AI are brought up in board rooms, executive meetings, and management discussions.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

The increased level of accuracy and level of detail (context) that it has grown to support—which means the answer/output can be much more tailored and specific to a need. I can see now why prompt engineering is really a specialized skill.

Sean Stein Smith

Accounting professor, Lehman College
Stein Smith-Sean-Lehmann College 2022
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

My perception of AI has evolved significantly during 2024, and this is due both to the improvements in available AI products, as well as the profession itself. Notably, the breadth of knowledge and subject matter areas that AI can now assist with is nearly universal, spanning accounting, audit, tax and advisory services. While still not 100% accurate or applicable, the results are impressive and allow CPAs (including myself) to draft, edit and work with the AI generated content to improve the end result. CPA firms have also embraced AI more substantially in 2024 than previously, as the profession has seemed to realize AI-augmented services can address several concerns facing the profession. Specifically, while the pipeline and retention issues remain significant headwinds, integrating AI into firm operations and client services allows firms (and the employees therein) to delegate more research and repetitive tasks, and focus on client relationship building as well as creating new revenue opportunities.

In short, the continued improvements in AI and the growing acceptance of AI practice by firms has made my perception change in a positive manner this past year.

What scares you the most about AI today? What is your biggest concern?

Two things about AI keep me up at night, and they are both related to the pipeline and the future of the profession. First, as AI takes over more of the routine, repetitive, and entry-level tasks and analysis there is a risk that some of the knowledge — and ability to explain financial information to clients — will be diminished. Second is the potential that as clients and businesses at large realize the abilities of AI that more clients might decide to completely delegate and outsource accounting and tax preparation to AI agents.

Both possibilities are fortunately able to be addressed with proactive leadership and planning. Employees at all levels, but especially newer higher, should be able to answer questions and explain financial analytics on the stop; that is what clients will continue to expect. In terms of AI outsourcing the fact is that such events will happen, but it is up to professionals to highlight that AI 1) does not always take into account the facts and circumstances of clients, 2) may be working with incomplete or incorrect data, and 3) might not be aware of the exceptions and workarounds that exist, but are not explicitly included in the tax code.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

I would say that top thing or change in the AI space, both as it relates to accounting AI and AI more generally, is the breadth of areas that GenAI bots and programs can now interact with humans on. Previous iterations of GenAI applications could only provide answers to certain questions phrased in the correct way, had issues with hallucination especially for technical subjects, and often were working with incomplete or outdated information.

The speed with which these issues have (mainly) be dealt with illustrates to me just how transformative AI is going to be for business and society at large. The pace of change can seem intimidating, but it also has been truly impressive to see how quickly the capabilities of AI have advanced this year.

Vsu Subramanian

SVP of content engineering and head of AI, Avalara
Vsu Subramanian.jpg
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

I think there is more awareness and more interest to see how AI can be used in accounting, especially in starting to use ChatGPT-like tools to do some of the mundane tasks. I think this represents a step forward and creates familiarity, and a realistic view, of what AI can and cannot do.

What scares you the most about AI today? What is your biggest concern?

Using AI responsibly is still a concern. Data privacy, governance, ethical use of data, and other considerations should be evaluated in the use of AI. These are still necessary steps anytime data is used, and we shouldn't forget these when using the data for AI.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

I've been impressed with rapid progress in Large Language Models, Visual Language Models and Small Language Models. As well as with the costs of Generative AI continuing to drop, and the emergence of the ability to fine tune these highly capable models.

In accounting, I see many products bringing AI into accounting solutions, but my impression is that we're still in the early stages.

Eyal Shinar

Co-founder and CEO, Black Ore
Eyal Shinar
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

This year has reinforced my perception of AI in accounting in two important ways. First, it has reinforced that AI is a powerful component, but not a complete solution, for our profession. What's become clearer in 2024 is the fundamental mismatch between how AI and accounting work: AI operates on probability, achieving increasingly impressive accuracy rates that approach — but never quite reach — 100%. Meanwhile, accounting demands deterministic precision; it operates on established rules and logic where even a 1% error in tax calculations or financial statements isn't acceptable to clients or regulators.

This fundamental understanding of AI's probabilistic nature has, in turn, shaped how the profession views AI's role. The anxiety of 2023 about AI replacing accountants has evolved into a more mature understanding in 2024. There's now widespread recognition that AI is a sophisticated tool that augments rather than replaces human expertise — freeing accountants to focus on strategic advisory work like tax planning where human judgment and experience are irreplaceable.

What scares you the most about AI today? What is your biggest concern?

What concerns me most is the accelerating gap between AI-adopting firms and those falling behind. It's creating a compounding effect in both directions that could reshape our profession faster than many realize.

Firms slow to adopt may enter a downward spiral: manual workloads lead to burnout and attrition, especially among younger talent who expect modern tools. Each departure increases burden on remaining staff and disrupts client relationships. Meanwhile, AI-forward firms create virtuous cycles: reduced manual work means engaged employees spending more time on strategic work like tax planning. This attracts talent, enables growth and funds continued investments in growth. Some firms are already handling 20% more clients with the same headcount.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

What's impressed me most is the dramatic leap in AI's fundamental capabilities — particularly the roughly 10x improvement in speed and cost efficiency of large language models. What cost $1 and took a second last year now costs pennies and happens in milliseconds.

Real-time processing is transforming how we interact with financial data — extracting insights from hundred-page contracts or analyzing a quarter's worth of transactions now happens instantly during client meetings. This isn't just faster, it's qualitatively different: we can explore implications and scenarios with clients in real-time, leading to deeper insights and better decision-making.

What's particularly striking is how these improvements compound each other — when every query costs 90% less, the depth and breadth of analysis we can provide fundamentally changes.

Prasad Sristi

Chief AI officer, Ascend
Prasad Sristi
How has your perception or impression of AI in accounting changed from the beginning of this year to now?  

AI in accounting is transitioning from excitement and initial forays into serious investments. You are seeing the focus change from efficiency gains to strategic transformation. You know firms are serious about this transformation when you see AI talent additions to their senior leadership roles. For example, Ascend brought me in because they truly prioritize the value of this transformation for our accountants and our clients. Investing in AI is a big part of Ascend's plan to revitalize the accounting industry.  

What scares you the most about AI today? What is your biggest concern?  

With all the interest in AI, I hope that we, as the innovation community, ensure we innovate responsibly with AI. As a community, it's important that we build meaningful solutions that truly make our customers' lives better and we stay away from driving investment towards pure hype. We all stand to lose if we build a bubble that gets popped. It is important that our innovation adds true value and builds trust with our users and the accounting community. We have a unique opportunity to create work-life balance in the accounting profession, empower our accountants to spend more time with their clients, and take away undifferentiated heavy lifting work from them. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.  

I'm quite impressed with the level of reasoning in the latest models. We are now talking about models that assess multiple possibilities and select the most appropriate response. I'm also impressed by how the top AI vendors are launching features that help the developer community. Developer friendly features such as function calling and outputting structured responses accelerate innovation with large-language models. Lastly, I'm impressed by how quickly prices have come down for the large language models. I think this is the fastest pace of price reduction for any new technology ever. 

Ben Wen

CEO and co-founder, Tallyfor
Ben Wen, Tallyfor
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

In 2024, AI tools for prosaic use cases like data extraction, analysis and categorization have superseded proof of concepts and broad marketing claims. I feel that the industry is learning how to shape the general AI tools like LLMs to be better fit for purpose for accounting. Personally I was optimistic at the beginning of the year, skeptical in the middle as I became more involved in an AI project, but am ending the year ever more optimistic about the impact of AI in accounting, saving time and reducing stress. 

What scares you the most about AI today? What is your biggest concern?

Generally, I do worry about the impact of AI in the hands of malicious users like security attacks and phishing. I am concerned that some borderline ethical use cases will cause real harm: physical, economic, or emotional.  I am a believer in Nick Bostrom's framing that we face "philosophy with a deadline," meaning that the ethical and existential impact of AI is coming and we better figure it out.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

For accounting, just the raw capability of the new public LLM models like GPT-o1 with good prompting is eye-opening. Outside of accounting, the programming tool Cursor is very impressive. It can understand more than small snippets of code, grasping large architectural goals.

David Wood

Accounting professor, Bringham Young University
Wood-David-Brigham Young
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

I believe much more now that GenAI, and the future developments related to it, will prove far more disruptive than I thought at the beginning of the year. We are starting to see significant, real-world use cases of this tech having a significant impact. The biggest inhibitor to progress at this point is not the technology, but waiting for people to start using and adopting it. We have significant work to convince people that it is OK to use, and not "cheating." As we do this, more and more will question previously performed tasks and processes and we should see increasing improvements in efficiency and effectiveness across an increasing set of outcomes.

What scares you the most about AI today? What is your biggest concern?

I continue to see the good and the bad about GenAI. There are obvious risks and people are going to use it for very bad things. I am concerned that some will try to over-regulate the technology because of this potential for bad things. I believe in punishing wrong doers, but I would hope we can continue to let those of us who want to use it for good to continue developing and exploring its potential. A second concern is that not enough people are trying it and learning about it, or, they have one experience where it doesn't work how they want it to and quit learning and exploring. It takes some work to figure out the positive and negative use cases, but once you do, it changes so much of what you can do.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

Two of my research projects amazed me. With my coauthors we tested if using GenAI could enhance auditor's fraud risk assessments and found that an intern with ChatGPT outperformed professionals with an average of 13 years of experience. Then, when we had ChatGPT perform the task without human help and it outperformed even humans using it…and then we used the reasoning models, and it did even better. Second, with my colleague Brant Christensen, we built a tool that could help perform some of the review process for audit and tax work papers and it absolutely amazed us at how well it worked. It found errors that we didn't mean to put in the workpapers. Each time we ran it, we were amazed at the things it could pick up in terms of finding poor judgments, poor documentation, formula errors, etc. GenAI's abilities are real.

Joe Woodard

Founder and CEO, Woodard
Joe Woodard.png
Joe Woodard
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the beginning of the year, I saw AI as an underinformed and often incorrect chat bot — a novelty with extreme potential. As the year progressed, I perceived AI as a way for developers to enhance their solutions. Now, at the end of the year, I'm seeing practical applications for accounting firms, including financial analytics and process automation. 

What scares you the most about AI today? What is your biggest concern?

Presently, my chief concern is about the reach of AI and how it is difficult — and increasingly infeasible — to regulate, given the level of innovation in the space, the escalating pace of development, and both the corporate and international competitive landscapes. Assuming we can find a way to regulate the development of AI — while maintaining national security — I'm also concerned about the accuracy of AI and how well-meaning individuals will blindly accept wrong or biased information and how bad actors and propogandists will increasingly leverage the technology. In the longer term, I'm concerned about AI becoming autonomous, even nefarious. 

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall.

Within accountancy I'm amazed to see that AI is, with the assistance of humans, performing sophisticated functions of accountancy like income tax preparation and income tax analysis (e.g., Intuit's "Done-for-You" platform), month-end close processes (e.g., Truewind and Botkeeper), and even end-to-end monthly bookkeeping cycles (e.g., Digits). As a larger category, I'm surprised to see how accurate, timely and capable AI bots like Chatbot have become, including the ability to "read" documents and spreadsheets, analyze information, and even create unique works of art.  

Carmel Wynkoop

Partner-in-charge of AI, analytics and automation, Armanino
Carmel Wynkoop, Armanino
Photo Credit - Robert Houser
How has your perception or impression of AI in accounting changed from the beginning of this year to now?

At the start of the year, AI felt like a promising tool for automation and efficiency. Now, it's clear that AI can also drive strategic insights, uncover trends and enhance decision-making, fundamentally reshaping the role of accountants. It also felt like AI was going to solve everything — I think after a year of usage we now understand better that we can use AI to solve the little mundane tasks that make us inefficient — thus leaving the harder and more intellectual tasks for us to solve.

What scares you the most about AI today? What is your biggest concern? 

AI's rapid evolution could lead to unchecked misuse, such as financial fraud or creating systemic biases, before we establish strong safeguards. I think we will be focused on safeguards/security, ethical use and data in the net year.

What's impressed you the most about AI this year? What really got your attention? Both in terms of accounting and overall. 

AI's ability to process unstructured data, like invoices or contracts, into actionable insights is transformative for accounting. Overall, generative AI's creative problem-solving is pretty phenomenal.
MORE FROM ACCOUNTING TODAY