ICAEW.com works better with JavaScript enabled.

AI: The next big thing in accountancy or the end of the profession as we know it


Published: 28 Mar 2023 Update History

Exclusive content
Access to our exclusive resources is for specific groups of students, users, subscribers and members.

Is Artificial Intelligence (AI) the next big thing in accountancy or the end of the profession as we know it? That's the question on the minds of many financial professionals as they grapple with the rise of AI.

Imagine a world where accountants work smarter, not harder – where software not only records transactions, but also analyses them, identifying trends and opportunities that might otherwise go unnoticed. Although this may sound like something out of a sci-fi movie, AI is making this vision a reality, transforming traditional processes and paving the way for exciting new innovations.

I’ve previously written about AI: What is it really?, and helpfully there is no universally agreed definition of AI, but broadly it is a field of computing focused around performing tasks that require human intelligence. With the ability to automate repetitive tasks, analyse large datasets, and provide valuable insights, AI is revolutionising the future of accountancy. Whether it's predicting financial performance or identifying potential risks, AI is undoubtedly changing the financial sector in its ability to automate, to draw on broad information sources, and provide new and innovative insights.

While generative AI solutions such as ChatGPT have recently been in the headlines, AI is about far more than this, so here we’re exploring some of the wider opportunities and challenges that artificial intelligence presents the accounting and finance industry.

There are several types of AI that are particularly useful in accountancy, including Machine Learning, Natural Language Processing and Robotic Process Automation.

The Machine Learning Accountant

Machine learning, as explained simply in “ML:What is it really?” is a subset of AI that involves the use of algorithms and statistical models to identify and predict patterns in large datasets, and in turn evolve and adapt to provide ever-improving results.

Accountants could use machine learning to analyse financial data and predict trends or identify potential risks, for example:

  • Credit control - by predicting the likelihood of a customer defaulting on a loan, it helps accountants in making informed decisions about lending risk. Historical and real-time data can be analysed and patterns identified, which can help inform financial planning and decision-making.
  • Fraud detection - through analysing large amounts of data and drawing links that go beyond the capabilities of typical human analysis, machine learning models can pinpoint potentially fraudulent activity and highlight it for further scrutiny, enabling accountants to tackle fraud risk in a proactive and targeted manner.
  • Budgeting and forecasting – ML models can analyse historical financial and non-financial data and predict future trends, enabling accountants to create more accurate budgets and forecasts.
  • Tax automation - compliance processes can be automated, ensuring that financial statements are accurate and comply with applicable tax regulations. For example, machine learning algorithms can automatically identify potential tax liabilities or deductions, reducing the risk of errors or omissions in tax filings. This was even demonstrated in the recent GPT-4 announcement, although a word of caution, some commentators have noted that the ‘answer’ was actually incorrect!

Deciphering Financial Data with NLP

Natural Language Processing (NLP) is another type of AI that can be used in accountancy. Essentially, NLP allows computers to understand human language by teaching machines to recognize the meaning behind words and phrases. This understanding can be used to interact with humans in a more natural way. NLP involves the use of computer algorithms to analyse, understand, and interpret human language, including text-based unstructured data and the spoken word.

Accountants can use NLP to automatically extract information from financial documents such as invoices and contracts, making it easier for accountants to process this information quickly and accurately. NLP can also be used to streamline auditing processes, improve customer service, enhance compliance, and manage risk more effectively.

  • Streamline auditing processes - by using NLP to automatically analyse and identify anomalies in financial statements, accountants can reduce the need for manual reviews and improve the efficiency and accuracy of auditing processes. NLP can help draw linkages between structured data such as trial balances, and unstructured data such as financial statement notes to support review processes.
  • Compliance monitoring - NLP can also be used in accountancy to automatically monitor and analyse regulatory filings and other compliance-related documents, ensuring that organisations comply with legal requirements and avoid penalties.
  • Risk management – tools that leverage NLP can help identify risks and threats to an organisation. By analysing vast amounts of unstructured information and identifying patterns and anomalies, NLP algorithms can provide valuable insights that accountants can use to enhance an organisation's overall security. It can also help manage contractual risk by extracting key data points from long, unstructured contracts.
  • Improve customer service - by analysing customer feedback and sentiment using NLP, teams can gain valuable insights into customer and client preferences, concerns, and behaviour. This information can then be used to enhance service offerings.

RPA: The Future of Accounting Efficiency

Robotic Process Automation (RPA) uses software robots, also known as "bots," to automate mundane tasks such as data entry, report generation, and other repetitive processes that would otherwise need human intervention. RPA technology is programmed to imitate the actions of a human user, enabling it to navigate and communicate with computer systems in the same way as a person would. In combination with AI, RPA can deliver tangible, repeatable outputs leveraging a degree of reasoning. For example:

  • RPA account reconciliation – RPA can be used to compare two or more data sources and identify similarities or differences between them. With the addition of AI it can remove some of the challenges of matching different data sources to each other (e.g. bank transaction descriptions to supplier details in finance systems), saving time and reducing errors.< /li>
  • Audit automation - RPA can be used by accountants to automate audit preparation tasks by automatically collecting data and compiling reports, with generative AI providing contextual narrative.
  • Data extraction – RPA and AI can be used to extract data from invoices and apply it directly in finance systems, reducing manual data entry.
  • Data management – accuracy and quality can be improved with the use of RPA and AI, which can automatically manage and organise data. For example, tools can be used to update customer records and consolidate financial data from multiple sources, or identify duplicates and ensure a ‘single version of truth’.

From Innovation to Isolation

Although AI offers a range of benefits for accountants, it's important to also consider its potential drawbacks. Some of these include concerns around data privacy and security, the possibility of job losses, the risk of bias, and the need for upskilling.

 Data privacy and security
 With AI-powered tools, accountants can process vast amounts of sensitive financial data, which could potentially be exposed to cyber threats. This gives rise to the risk of data breaches and privacy concerns. Organisations must ensure that their AI systems are secure and comply with data privacy regulations to mitigate against these risks.
 AI Bias
 AI solutions are really only as good as the data they are trained on. Therefore, it is of upmost importance that the data is as unbiased as possible, as otherwise algorithms will often accentuate any inherent bias, rather than overcome it.
Job displacement and the need for upskilling
 Another challenge associated with AI in accountancy is the possibility of job displacement. As AI becomes more advanced, it's possible that some routine tasks currently performed by accountants could be automated, leading to job losses. It's important to note that this is not the first time technology has impacted the accounting profession. In the past, the introduction of computers and accounting software displaced some bookkeeping roles. However, it also created new opportunities for accountants to focus on higher-value tasks, such as providing financial advice and strategic planning.

As AI technology continues to shape the accounting profession, accountants must acquire new skills to stay relevant. Some of these skills may include data analysis and interpretation, programming and coding, proficiency in cloud-based software, and expertise in specific AI-powered solutions.

To minimise the impact of job displacement, it's crucial that organisations invest in upskilling programs to help their workforce develop new skills and knowledge necessary to stay competitive in the job market and meet the demands of a rapidly changing profession.

Bean Counters to Bot Builders: The Future of Accountancy

Accounting has a long history, dating back many centuries. But as technology has progressed, so has the profession. The emergence of computers in the mid-20th century allowed for faster and more accurate processing of financial data, leading to the widespread adoption of computerised accounting systems in the 1980s and 1990s. And so, recently, new technologies like AI will continue to cause the accounting sector to evolve. With the ability to process and analyse vast, disperate amounts of data, accountants can make more informed decisions and be more forward-looking, helping stakeholders understand what might happen, rather than what has happened.

There are more use cases than we can possibly cover in a single article. Virtual assistants or ‘chatbots’ can handle most routine inquiries, while AI-powered advisors may soon start to provide more tailored, personalised recommendations based on a far greater wealth of information than ever before. And even blockchain technology brings AI opportunities, as the volume of data grows ever further and advanced tools are required to analyse and interpret it.

Ultimately, the future of accounting will depend on how we choose to use and regulate AI. By embracing the potential benefits of AI while also being mindful of the risks, we can create a future where humans and machines work together to create a more efficient and effective accounting profession.

About the author

Monica Odysseos