AI revolution testing ethical standards in accountancy

AI revolution testing ethical standards in accountancy

Although AI is expanding the capacity to process and analyse financial information, its use raises fundamental questions about the limits of automated outputs, the role of human judgement and the accountability of finance professionals.

The growing integration of AI into financial analysis and decision-making is raising complex ethical questions for the accountancy profession.

The challenge, as set out in the Association of Chartered Certified Accountants’ (ACCA) Navigating Ethics for Trust report, published on October 14, is to maintain trust as the tools used to produce and assess financial information evolve.

Drawing on responses from 1,178 finance and accountancy professionals across 135 countries, alongside roundtable discussions, the report examines how emerging technologies interact with established ethical concerns, including leadership, organisational culture and fraud. Its findings also raise questions about whether professionals have sufficient practical support to apply longstanding ethical principles in increasingly complex working environments.

The survey found that 61 per cent of respondents ranked the use of AI among their top three ethical challenges over the past three years, up from 32 per cent in the 2024 survey. Ethical leadership was cited by 43 per cent of respondents, followed by ethical culture at 39 per cent, placing them second and third among the most frequently identified challenges.

Other concerns also gained prominence: the proportion citing the geopolitical environment rose from 25 per cent in 2024 to 34 per cent in 2026, while the share identifying fraud increased from 24 per cent to 33 per cent.

ACCA’s research point to a changing set of ethical pressures on the profession, with technology gaining importance alongside longstanding concerns about leadership, organisational culture, and misconduct. According to the report, many professionals faced difficulties applying ethical principles at work.

Half of respondents said ethical dilemmas had become harder to resolve over the past three years, while 53 per cent had witnessed unethical behaviour during the same period. A further 22 per cent said they had been pressured to act against ethical standards.

The share of respondents who found ethical dilemmas harder to resolve declined from 64 per cent in 2024 to 50 per cent in 2026, while the proportion answering “don’t know” or “not applicable” rose from 14 per cent to 25 per cent. These figures represent two notable changes in how respondents assessed the difficulty of resolving ethical dilemmas over the period.

Sarah Lane, head of Ethics and Assurance at ACCA, said the accountancy and finance profession had generally maintained its ethical foundations and ability to adapt, but professionals needed further support in applying ethical principles in complex working environments.

“Finance professionals are not asking for more rules, because the fundamental principles remain their ethical anchor. What they need are practical tools, support, and guidance to apply these principles in an increasingly complex world,” said Lane.

Professional ethics

According to the ACCA survey, 72 per cent of respondents said AI tools had improved the quality and efficiency of their work, while the same proportion reported that understanding AI helped them use it more responsibly. A further 89 per cent said they applied professional scepticism and ethical judgement when using the technology.

Despite these reported benefits, AI adoption presents challenges involving verification, oversight and accountability. ACCA identifies over-reliance on automated outputs, inadequate checks, information security concerns and limited transparency among the risks facing finance and accountancy professionals. Weak organisational governance and unclear responsibility for decisions informed by AI-generated results add to these concerns.

The reliability of AI-assisted analysis depends on the quality of input data, the assumptions underpinning the results and their suitability for the intended purpose. In financial reporting, errors or unsupported assumptions can affect the information presented to stakeholders, making the basis for AI-generated outputs a relevant consideration in their use.

Beyond technology-related risks, the survey examined the factors respondents associated with trust in the accountancy and finance profession. Strong ethical standards ranked highest among the top three factors, selected by 46 per cent of respondents, followed by transparency at 38 per cent and independence at 36 per cent. When asked to identify the single most influential factor, 18 per cent chose ethical standards, while transparency and independence each received 15 per cent.

Sustainability was another area covered by the survey, although the proportion of respondents ranking it among their top three ethical challenges fell from 33 per cent in 2024 to 21 per cent in 2026. In sustainability reporting, ethical risks can arise from incomplete data, selective use of indicators and environmental claims that are not adequately supported by evidence. Greenwashing becomes a concern when disclosures convey a more favourable impression of environmental performance than the available data justify.

For professionals involved in sustainability reporting, these issues centre on the quality and consistency of disclosed information and the evidence supporting corporate claims. The scope of review depends on the nature of the engagement and the standards applicable to it.

Ethics into practice

The ethical challenges outlined in the report point to the need for practical guidance that helps professionals interpret and apply ethical principles in complex situations. To support this, ACCA provides a toolkit featuring a 10-step practical process, alongside guidance on applying the International Code of Ethics for Professional Accountants issued by the International Ethics Standards Board for Accountants.

ACCA’s toolkit sets out a 10-step process for addressing ethical dilemmas, from identifying the issues involved to determining an appropriate course of action. It provides a framework for considering how established ethical principles apply in practice, particularly when technological developments, organisational expectations or commercial pressures create competing demands.

The report also examines the role of education in developing ethical competence, and Sarah Lane said ethics had long been part of ACCA’s training programme.

“The organisation plans to introduce a redesigned programme in 2027, with ethics integrated throughout the curriculum. The change reflects the need to consider ethical questions across different areas of professional practice, rather than treating them as a separate topic,” Lane said.

Understanding ethical principles, however, does not automatically resolve the dilemmas that arise in practice. Professionals may need to interpret existing standards in circumstances that those standards do not explicitly address, weighing different obligations before reaching a decision. The report’s discussion of training and practical guidance points to two related aspects of ethical competence: knowing the principles and being able to apply them when circumstances are less clear-cut.

Lane added that ethics had long been a core component of ACCA’s training programme. The organisation plans to introduce a redesigned programme in 2027, with ethical considerations integrated throughout the curriculum rather than confined to a standalone subject.

“The approach is intended to prepare future professionals to navigate ethical dilemmas arising in different areas of their work as technology, workplace practices and business expectations continue to change,” she said.

Melanie Proffitt, president, ACCA Global

AI can handle repetitive tasks, process large volumes of data and assist with analysis. Yet the quality of its output depends on factors such as the data used, the assumptions built into a system and how its findings are interpreted. Faster processing does not necessarily produce more reliable conclusions.

For accountants, this means assessing AI-generated outputs in the context in which they will be used, rather than accepting them at face value. This involves checking the supporting evidence, considering potential errors and applying professional knowledge to determine whether the results are appropriate for the task.

The use of AI also raises questions about the allocation of responsibility. Even when a system performs part of the work, accountants and businesses must establish who reviews its output and takes responsibility for decisions based on it. How these responsibilities are defined will become increasingly relevant as AI is incorporated into financial and accounting processes.

Ren Varma, regional head Mainland Southeast Asia, ACCA

Southeast Asia is adopting AI in finance at a pace that is testing existing governance and ethical guidance. In the survey, 61 per cent of respondents identified AI use as a leading ethical challenge, nearly double the proportion recorded two years earlier. The findings point to a growing gap between the uptake of the technology and the frameworks needed to address questions of accountability and professional conduct.

For finance professionals in Vietnam and across the region, a key issue is how responsibility should be assigned when decisions rely on AI-generated outputs. Existing ethical principles provide a foundation, but their application becomes more complex as technology changes the way financial information is produced and assessed. The focus is therefore shifting towards how established standards can be applied in practice, including how AI-generated results are reviewed and who remains accountable for decisions based on them.

VIR