Change is inevitable. As accountants, we tend to be conservative; if anything, reactive to change. We are typically slow to adapt as we analyze before acting. Being slow, methodical and detailed are characteristics that seem to attract accountants to the profession, they’re certainly characteristics that are part of the professional training. However, change comes regardless of whether we act; it is inevitable. A change that has come recently and is poised to make an impact to every facet of life, including accounting, is the dawn of Artificial Intelligence (“AI”) and Machine Learning (“ML”). This change begs some interesting questions that have begun speculation including, “How will AI and ML impact the accounting profession?” and “What are you doing to prepare for it?”.
I thought it appropriate to use artificial intelligence in my analysis of artificial intelligence to illustrate the pervasiveness and accessibility of this relatively new technology.
Please note that Microsoft Copilot was used below, and I have included the credit citation, as given by Copilot, where applicable. As is a typical AI disclaimer, beware of inaccuracies. The initial questions belong to me, but the responses were generated by AI. Please review, and following the interaction with the AI tool, I will provide some additional thoughts.
You said: How is AI and ML affecting the accounting industry?
Copilot said: AI and ML are significantly transforming the accounting industry in several ways:
- Automation of Routine Tasks: AI-powered systems can handle repetitive tasks such as data entry, invoice processing, and transaction categorization. This reduces the risk of human error and frees up accountants to focus on more strategic activities[1].
- Enhanced Data Analysis: Machine learning algorithms can analyze large datasets to identify patterns and anomalies. This helps in detecting fraud, forecasting trends, and providing deeper insights into financial data[2].
- Real-Time Financial Monitoring: AI systems can process and analyze financial data in real-time, allowing for more timely decision-making and improved financial management[1].
- Improved Compliance and Accuracy: AI can help ensure compliance with regulatory requirements by continuously monitoring transactions and flagging any irregularities. This enhances the accuracy and reliability of financial reporting[1].
- Skill Development: As AI and ML become more integrated into accounting, professionals need to develop new skills in data analytics, AI tools, and technology management. This shift is leading to a more tech-savvy workforce[2].
Overall, AI and ML are reshaping the accounting landscape by making processes more efficient, accurate, and insightful.
[1]: Focus People [2]: IMA
References
[1] 2024 and Beyond: The Impact of AI on the Future of Accounting
[2] The Impact of Artificial Intelligence on Accounting and Finance
You said: Can you be more specific as it relates to public accounting firms?
Copilot said: Certainly! AI and ML are having a profound impact on public accounting firms in several specific ways:
- Audit Automation: AI tools can automate many aspects of the audit process, such as data extraction, analysis, and anomaly detection. This not only speeds up the audit process but also enhances accuracy by reducing human error[1].
- Tax Compliance: AI can help public accounting firms manage complex tax regulations by automating the preparation and filing of tax returns. Machine learning algorithms can also identify potential tax savings and ensure compliance with ever-changing tax laws[2].
- Client Advisory Services: AI-driven analytics provide deeper insights into clients’ financial data, enabling firms to offer more strategic advisory services. This includes predictive analytics for financial planning and risk management[3].
- Fraud Detection: Machine learning models can analyze vast amounts of transaction data to identify unusual patterns that may indicate fraud. This proactive approach helps firms detect and prevent fraudulent activities more effectively[1].
- Efficiency and Cost Reduction: By automating routine tasks, AI allows accountants to focus on higher-value activities, improving overall efficiency and reducing operational costs. This is particularly beneficial for public accounting firms that handle large volumes of transactions and data[2].
- Enhanced Client Experience: AI can streamline client interactions through chatbots and automated customer service, providing quick responses to common queries and improving overall client satisfaction[3].
These advancements are helping public accounting firms stay competitive and deliver more value to their clients.
[1]: Airbase [2]: ICPAS [3]: Peakflo
References
[1] Accounting and AI: How AI & ML Impact Finance Teams – Airbase
[2] The Transformative Impact of AI on Accounting – ICPAS
You said: Can you talk about the limitations of AI and ML as it relates to the above? For example, there are concerns that AI and ML technology will replace jobs currently occupied by people; what is it that people are still better at than machines?
Copilot said: Absolutely, while AI and ML bring numerous benefits to public accounting firms, there are several limitations and areas where human expertise remains crucial:
- Human Judgment and Expertise: AI can process and analyze data at an unprecedented scale, but it lacks the nuanced judgment and ethical considerations that human accountants bring. Complex decision-making, especially in ambiguous situations, still requires human intuition and experience[1].
- Contextual Understanding: AI may struggle with understanding the context behind certain financial transactions or decisions. Humans are better at interpreting the subtleties and nuances that can affect financial outcomes[1].
- Data Security and Privacy: Implementing AI involves handling sensitive financial data, which raises concerns about data security and privacy. Ensuring that AI systems are secure and compliant with regulations is a significant challenge[2].
- Job Displacement Concerns: There is a valid concern that AI and ML could replace jobs currently occupied by people. However, while AI can automate routine tasks, it also creates opportunities for accountants to focus on higher-value activities such as strategic planning and advisory services[3].
- Adaptability and Learning: AI systems require continuous updates and training to adapt to new regulations and financial practices. Humans, on the other hand, can quickly adapt to changes and apply their knowledge in real-time[3].
- Ethical Considerations: AI lacks the ability to make ethical decisions. Human accountants are essential for ensuring that financial practices adhere to ethical standards and for making decisions that consider the broader impact on stakeholders[1].
In summary, while AI and ML are powerful tools that enhance efficiency and accuracy
in public accounting, human accountants remain indispensable for their judgment, contextual understanding, and ethical considerations.
[1]: Canopy [2]: Trullion [3]: Space Coast Daily
References
[1] How AI is Used in Accounting: Capabilities and Limitations
[2] AI in Accounting: The Benefits & Challenges – Trullion
[3] AI Implementation in Accounting: Use Cases, Challenges and Trends
Final Thoughts
Copilot uses Large Language Model (“LLM”) algorithms using neural network techniques to process and understand vast amounts of data enabling it to generate the above responses. There are types of models other than LLMs, that analyze types of data other than language, like numerical data, even images.
But where does this underlying data come from? Who has designed the algorithms that run the engines behind these systems? Are there biases included in their design? Who owns it? Where can we find it? How will it be regulated? How will it be monetized?
Carl Sagan once said, “We have arranged a society based on science and technology, in which nobody understands science and technology… this combustible mixture of ignorance and power, sooner or later, is going to blow up in our faces… Who is running the science and technology, in a democracy, if the people don’t know anything about it? …If we are not able to skeptically interrogate those who tell us that something is true, to be skeptical of those in authority, then we’re up for grabs for the next charlatan… who comes ambling along. (Per Jefferson), it wasn’t enough to enshrine some rights in a constitution… the people had to be educated, and they had to practice their skepticism and their education, otherwise, we don’t run the government, the government runs us.”
How can you question the truth if you have no underlying knowledge about that truth’s subject matter? The data and algorithms that underly this technology are proprietary. They are a black box to which we, the public, often have no real insight. To practice skepticism, we shouldn’t blindly trust this technology (or the people who design and control it).
AI and ML will be useful tools, and they will help us find efficiency and even new revenue streams. But I think accountants will remain relevant because someone must understand the subject matter to maintain the skepticism that Sagan refers to. We should not give away our agency to AI, and we should be skeptical of the underlying data and engines it uses. As accountants, to do that, we must understand how accounting works. But in the future, to remain in our role as trusted advisors and to effectively use these tools, we’ll also have to have at least a basic understanding of what neural network techniques are as well as other statistical models behind how AI and ML tools work. Perhaps we can have a hand in delaying that combustible mixture Sagan mentioned from blowing up in our faces for a little while longer.
Welcome to the age of Data Science!
Article written by: Scott Nabozniak CPA, CA