AI doesn’t just mean accountants will be increasingly working hand in glove with robots. Working methods and business models will also transform in what is increasingly called the fourth industrial revolution. With this will come a change in the traditional accountancy skills needed – whether you are an accountant in practice or business, the changes that AI technologies are bringing to most roles will be substantial.

Organisations have anticipated an accelerated shift in the division of existing work tasks between humans and machines in the years to 2022, according to the World Economic Forum’s The Future of Jobs Report 2018. Four years ago, an average of 71% of total task hours were performed by humans, compared to 29% by machines. By 2022 this average is expected to have shifted, with humans carrying out 58% of task hours compared with machines doing 42%.

The report says that by 2022, 62% of organisations’ information and data processing, and their information search and transmission tasks, will be performed by machines, compared to 46% in 2018. That evolution has already been gaining pace. Much of the repetitive process-driven tasks are already being carried out increasingly by machines within the profession.

As AI’s influence in the working world spreads, accountants will have to diversify their skills. These developments will fall into two categories, argues Richard Anning, Head of the Tech Faculty at ICAEW.

First, accountants will need to understand at a basic level how AI works and how to work with AI specialists. “This doesn’t mean accountants need to become technology specialists themselves,” Anning explains. “They will, however, need more technology knowledge and skills than has perhaps been needed in the past to be intelligent and responsible users of the tech.”

This technical understanding is something that Lee Holloway, Tax Partner at Grant Thornton, sees as particularly important: “The ability to understand end-to-end processes and data transfer in order to audit the data sufficiently would be crucial from a tax perspective.”

The second part of the skill shift, says Anning, is slowly freeing up accountants to focus more on advisory, analysis and business partnering, as AI deals with more of the processing and transactional tasks. With these tasks removed for the most part, accountants can focus on developing their soft skills such as emotional intelligence, negotiation and persuasion skills, as well as sharpening their broader business knowledge. “Many of the members that I speak to highlight this shift as just as important, if not more so, than tech skills,” Anning says.

But this evolution of roles won’t happen overnight. Anning foresees a subtle but distinct transformation in roles over the coming years, with new jobs that are more at the nexus of tech and accounting. These could include roles focused on training machine-learning models, or on data governance, reviewing data quality, standards and ethics across all parts of the organisation. He also predicts a possible rise in roles that focus on reporting or assurance of different types of non-financial information, or assurance around the technologies itself.

The increased focus on advisory skills means that accountants will often find themselves acting as the broker between technical experts and clients. While people looking at their own communication skills often focus on delivery of ideas, Anning says good listening skills and an ability to understand the perspectives of different audiences is vital if accountants are to have meaningful conversations around AI. He also emphasises the importance of constructing a compelling narrative: “Storytelling and effective visualisation skills are examples of the kinds of softer skills that can help in this regard.”

Clive Bawden, founder of governance adviser BoardSecure, says the core skills will still lie in knowing what questions to ask your clients so that you have a thorough understanding of their objectives and which data will serve them best. According to Holloway, accountants who have undertaken some gritty work in the field will have far greater confidence conducting these advisory conversations. His advice is to get stuck in and do any work with machine learning to develop these skills.

Practitioners who will set themselves apart will be those who can formulate a clear roadmap for success with universal buy-in from all parties. “It’s about ensuring that the technical details can be translated into actions that everyone agrees and understands,” Holloway says.

There is no better way of increasing your knowledge and confidence in a subject than discussing it with the experts, which is why networking and educational events will be more important than ever. Anning is an advocate of upskilling within the profession and the Tech Faculty convenes a number of forums, such as the Mid-Market Tech Forum, where peers can effectively share their experiences.

“Learning from peers can be a useful way of understanding what’s going on in the real world, rather than just hearing all of the hype from the tech world,” says Anning. It will be vital to develop professional networks that include IT professionals and industry experts who are really using these AI tools and systems – a lot of people who talk about AI are trying to sell a tool without actually having implemented it in the real world, Bawden says.

Despite these exciting developments, the consensus seems to be that some immutable skills within the profession will always be important. One thing is for sure, says Holloway: “Humans will still need to interpret data and make use of it in the context of their business.”