Will AI replace tax consultants?

Over the last few months, conversations about the future of work have been dominated by attempting to define the role of generative AI tools like Chat GPT. The tax consultancy profession is no exception, and to try to assess the impact of AI in our field, KPMG has this month published a concise analysis: The use of Generative AI tools in the Tax Profession- After the initial hype- Fear, Foe or Friend

By René Schöb, Partner, Head of Tax and Legal, KPMG România

We are at an early stage but the key will be to combine AI and human skills

Apart from a few early adopters, most tax professionals are probably still playing with AI to try it out and see what it will and won’t do. Some will be enthusiasts while others will be sceptical and eager to point out its flaws. Others may be worried about their jobs. But trying to look through all the hype, fears and expectations, it seems clear that the way forward is to use AI to enhance the level of service which tax consultants can provide, taking advantage of the efficiency which it can bring, while also deploying human skills where these are most useful. The key will be to get the right balance, and if this can be achieved, the outcomes for tax consultants and the organisations they serve can only be positive.

Read more about

Ayn Rand Centre: 63% of businesses want to adopt artificial intelligence in three years

read more

How can AI be used to enhance the quality of tax consultancy?

The KPMG report gives several examples of how AI can help tax professionals in different areas of their work. Tax compliance nearly always involves the extraction of data, usually from multiple sources, then reconciliation of data and calculation, followed by the final outcome, normally in the form of a tax return or compliance filing. AI has the potential to streamline and render obsolete many of the very time ­consuming aspects of this process, leading to significant efficiency gains. In Tax controversy there is a lot of scope for using AI’s capacity for acquiring and sifting large amounts of information. For example if a submission is being prepared to a tax authority, on behalf of a taxpayer, Al can find relevant legislation, case law, or rulings and even anticipate the line which might be taken by the tax authorities. In tax litigation, AI can significantly speed up the process of preparing documentation and analysing the information provided by the opposing party.

Tax advisory, whether ongoing or related to a particular transaction, primarily focuses on opportunities and risk management. AI can analyse complex tax structures and hence support tax planning. It can quickly evaluate the tax implications of various business decisions and also enhance risk management through instant access to appreciated knowledge and insights.

AI can support knowledge management

Because of the highly technical nature of their work, tax professionals often struggle to communicate the impacts of complex technical tax matters to non-tax professionals. This can impede the effective integration of the tax department into the rest of the business. AI can help bridge this gap by generating simplified analyses and narratives for complex tax concepts and regulations, making them more accessible for non-tax professionals to understand. Hence it can aid the sharing of knowledge across the business, help staff in different departments to work more closely together and so contribute to better performance for the organisation as a whole.

A word of caution: Protecting confidentiality and ensuring accuracy

The ethical debates around AI are well known, but there are particular issues to address in relation to tax consultancy and business. For example, bearing in mind that AI “learns” and uses the information it assimilates, it is critical to protect confidentiality. As such, the Al tools used need to be (virtually) ring fenced in private environments, so that while the Al functionality may be available to its users, the queries or outputs from it do not leave the secure environment. Moreover, the possibility of errors must not be ignored. Consequently, ensuring that the source of information which is produced by Al is known, validated and tested, will be an increasingly important discipline.

So will we still need tax consultants?

The answer, of course, is yes. While AI can be expected to reduce some routine roles, for example in processing tax compliance, there are also likely to be newly emerging jobs, linked with a change in the nature of tax consulting services (whether internal or outsourced). The new roles are likely to focus on creating, testing, controlling and monitoring. Moreover, the human skills which tax consultants deploy in relations with clients – such as empathy, experience, judgment, communication and active listening- will be critical. Human knowledge and experience will continue to be valued as an essential complement to what AI provides. In short, the concept of trusted tax advisor is unlikely to disappear.