The paper Generative Artificial Intelligence in the Big 4 Auditor Adoption and Its Implications for Audit Quality by Yueqi Li and Sanjay Goel shows that GenAI is already changing how large audit firms work, but its effect on audit quality is mixed rather than purely positive. For internal audit leaders, audit committees, CFOs, and corporate governance professionals, the key message is clear: GenAI can improve speed, drafting, research, and issue spotting, yet it can also weaken skepticism, transparency, and control if it is used carelessly.
Based on interviews with 37 Big 4 audit professionals, the study finds that GenAI is most useful for routine and language heavy work such as writing memos, researching accounting standards, summarizing information, and preparing workpapers. Auditors also reported that the tools can help with planning, documentation, and communication, which can free up time for higher risk judgments. This is an important finding for Internal Audit because it suggests that GenAI should be positioned as an augmenting capability, not as a shortcut for core assurance work.
A particularly noteworthy insight is that GenAI adoption differs sharply by experience level. Junior auditors appear to benefit most from drafting support, basic explanations, and help starting workpapers, while seasoned auditors use GenAI more for alternative viewpoints, risk assessment support, and review enhancement. That distinction matters for governance because it shows that one technology policy will not fit every user group. Audit Committees and oversight bodies should expect firms to tailor controls, training, and usage rules to the maturity of the user, the task, and the risk of the engagement.
The study also highlights the sharp edge of GenAI in audit quality. Interviewees praised the technology for reducing some human error, improving efficiency, and surfacing overlooked risks, but they also warned about hallucinations, weak audit trails, confidentiality exposure, biased outputs, and overreliance on machine generated content. The most striking concern is that overuse could erode professional skepticism over time, especially if junior auditors begin relying on GenAI before they have built the judgment needed to challenge it. For internal audit functions, that means GenAI governance must include verification steps, clear documentation standards, and explicit rules on when human judgment must override the tool.
For corporate governance and supervisory bodies, the practical implication is that GenAI cannot simply be adopted as an efficiency project. It must be treated as a governed capability with controls over data quality, privacy, security, prompt discipline, output review, and accountability. The article also shows that adoption is shaped by social influence, management support, time pressure, and the perceived maturity of the firm’s technology environment. That means audit committees should ask not only whether GenAI is being used, but also whether there is enough training, whether controls are tested, and whether the firm can explain how the tool supports rather than replaces audit judgment.
Overall, the article offers a timely warning and a practical roadmap for assurance leaders. GenAI can raise audit quality when it is used to strengthen efficiency, broaden insight, and support better documentation, but it can also create new governance risks if firms assume that convenience equals reliability. The full article ‚Generative Artificial Intelligence in the Big 4 Auditor Adoption and Its Implications for Audit Quality‘ by Yueqi Li and Sanjay Goel is available here.