Why AI Can't Do Accounting (Yet)?

If you were a software engineer with a CPA license (or vice versa), how would you answer this question?

Despite ChatGPT passing the CPA exam and despite the potential for AI to automate much of the accounting processes, there are still some core reasons why AI is not (yet) capable of entirely replacing human accountants.

Why?

To understand this fully, one needs to learn how AI is created, how it learns on its own, and how it works.

Fundamentally, accounting is not solely about computations.

• AI algorithms can efficiently process and analyze vast financial data. But accounting always requires context and discretion to make informed conclusions from the financial data. For example, an AI system may identify a strange expense, but without additional information and scrutiny, it cannot determine whether that expense is legitimate or fraudulent.

• Accounting demands knowledge of business operations and strategies. While an AI system may generate financial statements and analyze financial data, it cannot provide strategic advice or make judgments about the direction of a business. Sure, ChatGPT can give some great answers, but that requires one to be an expert to ask the most accurate and relevant questions (and then apply an expert judgment to make sense of the answer). On the other hand, human accountants can provide valuable insights and guidance to their clients using their experience and expertise. The happiest of clients say this about their accountant:

• Accounting demands substantial communication and interpersonal abilities. Accountants must communicate intricate financial information to their clients in a straightforward and comprehensible manner. They must also establish client relationships and work effectively as part of a team. While AI systems can generate reports and data visualizations, they can lack the personal touch and communication skills that human accountants possess.

• Accounting requires a significant amount of judgment and discretion. Accountants must decide how to classify and report financial transactions and use their discretion to ensure that financial statements adhere to accounting standards and are accurate. While AI systems can assist with this process, they can lack the judgment and discretion required to make these decisions independently.

• Accounting is based on ethical considerations. Accountants must adhere to ethical principles and standards and prioritize their clients' and the public's interests (over their own). While AI systems can be programmed to (learn to) adhere to ethical guidelines, they can lack the ability to understand and apply these principles in a nuanced and context-specific manner. Unsupervised AI learning can further threaten this process as biases can creep in.

• Technical limitations hinder AI from entirely replacing human accountants. AI algorithms require substantial amounts of high-quality data to learn and improve, and there are often limitations on the availability, access to, and quality of financial data. AI algorithms may not yet be capable of handling all types of financial transactions and complexities.

These are just a handful of crucial reasons why AI can't yet fully replace accountants. To get a comprehensive, more profound understanding of what AI will find challenging to replace in accounting, get familiar with the fundamentals of AI (and no, it does not require any prior knowledge of AI to learn about AI.)

Don't underestimate AI.

If you are an accountant reading this and what is mentioned above is making you feel comfortable, you would love to learn more about AI.

Regardless of any (current) limitations, AI can play a vital role in accounting. AI can be utilized to significantly automate routine tasks like data entry and reconciliation, allowing accountants to focus on higher-level tasks that require human judgment and expertise. AI can also be utilized to analyze large datasets and identify patterns and insights that may be commercially unviable or manually very difficult to be immediately evident to human accountants.

Fully harnessing the benefits of AI in accounting would require a comprehensive understanding of both - the technical aspects of AI and the practical considerations of accounting.