The AI Audit Revolution: Less Grunt Work, More Trust
Auditing requires precision, professional judgment, and confidence in the information being reviewed. But let’s be honest. It also involves a lot of repetitive, time-consuming work.
In this episode, I talk with Steven Figner, CPA, CA, CEO and cofounder of Decision Point Advisors and president and CEO of Alethia Technologies. We discuss how AI agents can handle some of the audit grunt work while keeping experienced professionals firmly involved in the final review.
Steven shares how his team is using multiple AI agents to review and map controls, learn from human feedback, and complete work that once took hours in approximately 20 minutes.
We also discuss the part that can’t be ignored: protecting confidential client and financial information while using AI.
AI Should Remove Repetitive Work, Not Professional Judgment
One of the most useful applications of AI isn’t replacing experienced professionals. It’s removing the repetitive work that keeps them from using their experience.
Steven’s team is applying AI agents to the review of service organization control, or SOC, reports. These reports can be lengthy, detailed, and time-consuming to map against the controls required for a financial statement audit.
The system uses one AI model as the worker and another as the judge. A human professional then reviews and validates the final results.
That human review is important. The AI completes the initial analysis, but the professional is still responsible for confirming that the conclusions make sense.
What Once Took Hours Can Now Take Minutes
For the specific process Steven’s team is testing, the initial review has been reduced to approximately 20 minutes.
The savings don’t stop with the first review. When a team member corrects or disagrees with something, that feedback is added back into the system. The model learns from each engagement and becomes more accurate and efficient over time.
That means less time spent finding, organizing, and mapping information and more time spent thinking critically about what the information means.
Better Work Can Help You Keep Better People
There’s also a people benefit.
New professionals traditionally spend a large amount of time performing basic, repetitive audit tasks. Once they understand the process, that work often becomes tedious and unrewarding.
AI agents can handle more of that repetitive activity while allowing team members to focus on review, analysis, problem-solving, and professional judgment.
That creates more interesting work and gives developing professionals an opportunity to use their critical-thinking skills earlier in their careers.
It may also help firms retain good people who don’t want to spend their careers doing work that can now be automated.
Confidential Information Must Stay Confidential
Speed isn’t the only consideration when financial and audit information is involved.
SOC reports and other financial documents may contain private, restricted, or confidential information. Uploading that information into a public AI platform may violate confidentiality requirements or the report’s terms of use.
Steven’s team addressed that issue by running its AI models locally. The information doesn’t leave the organization or get sent to an outside cloud-based model.
This is an important lesson for every business using AI. Before uploading sensitive information, you need to know where that information is going, how it will be used, and who may be able to access it.
What Using an Audit Agent Could Look Like
The goal is to make the process straightforward for the accounting or audit team.
A user uploads information through a browser interface, often using a familiar Excel or CSV file. The AI agent analyzes the data and presents its findings. The professional then reviews, validates, and corrects the results when necessary.
The more the system is used and reviewed, the better it can become at identifying relevant information and reducing false positives and false negatives.
The Bigger Opportunity Is Trust
AI can help audits become faster and more scalable, but the larger issue is trust.
Businesses need confidence in their systems, transactions, controls, and data. As those systems become more complex, traditional audit approaches may become harder and more expensive to apply.
AI agents may give professionals a better way to review larger amounts of information without sacrificing oversight, privacy, or professional judgment.
The technology is still evolving, but the opportunity is clear: use AI to reduce the grunt work so your people can spend more time on the work that requires experience, judgment, and trust.
Listen to the episode, then choose one repetitive process in your business that could benefit from automation while keeping a knowledgeable person involved in the final decision.
Frequently Asked Questions
What are AI agents in auditing?
AI agents are specialized software tools that can complete specific audit tasks, such as reviewing documents, identifying controls, mapping information to established frameworks, and preparing results for human review.
Can AI agents replace human auditors?
AI agents can automate repetitive audit work, but experienced professionals are still needed to review the results, apply professional judgment, investigate exceptions, and accept responsibility for the final conclusions.
How can AI reduce the time required for an audit?
AI can review and organize large amounts of information much faster than a person performing the same repetitive steps manually. In the example discussed during this episode, an initial SOC report review was reduced to approximately 20 minutes.
How can accounting firms protect confidential data when using AI?
Firms should understand where their information is stored, whether it is shared with outside platforms, and whether it may be used to train external models. Local or privately hosted AI models may provide greater control over sensitive information.
How can AI improve employee retention in accounting firms?
By automating repetitive work, AI can allow team members to spend more time on analysis, review, problem-solving, and client service. That can make the work more challenging, rewarding, and professionally valuable.