The practice of auditing is more than a century old.What began as an accountant’s job is now practicedacross industries, with sophisticated focus onmeeting regulatory requirements or performingrisk-based audits. Although the types and methodsor models of auditing have evolved over time,the auditor community has faced significantproblems ranging fro.
Auditing will be a key governance mechanism to confirm that AI systems are designed and deployed in line with a company’s goals
But to create a risk-based audit plan specific to generative AI, Internal Audit must design and adopt new audit methodologies, new forms of supervision and new skill sets
For any AI program to be successful at solvingauditing problems, it needs to target the problemsof data and data sets. In this regard, the answer toall of the following questions must be “yes”:.
1) Is the origin or source of the data known?.
2) Are the data easily accessible?.
3) Are the data well rounded and reliable (i.e., is data integrity assured).
AI will have a far-reaching impact on the audit profession as well, given auditors’ need to provide AI assurance
Auditors should ask themselves whether organizations and audit teams are ready for the tough questions surrounding AI and the approach with which it is to be audited
Auditors must deal with a sea of information anddata presented in response to compliance andother areas. It often seems impossible toconsistently make sense out of audit samples. Thebiggest opportunity for RPA, ML and AI to work forauditing is to provide insights and intelligenceregarding the sea of data. These opportunitiesinclude the following: 1.
Figure 2summarizes the automation/AI scope foreach step of the audit workflow. It is evident that RPA, NLP and predictive analysisare some of the techniques that could bolster theway auditors approach audits.3, 4
Any audit program can be measured using thefollowing parameters to gauge itseffectiveness/success:.
1) Environment—The factors impacting the work of the internal audit function.
2) Output—The end results of the audit function.
3) Quality—The quality of end results.
4) Efficiency—The measure of output and quality of results vs. costs.
5) Impact—The impac.
There have been several applications of AI and ML inother fields such as anti-money laundering. Detecting fraudulent transactions, performing data qualitychecks, negative news screening and processinghave all been successfully automated via AI/MLtechniques. Implementing AI or ML for largemultinational corporate banks leads to big savings inmanual o.
As our clients embrace these new technologies, Deloitte’s Generative AI practice will serve to support clients in the development and deployment of new and innovative AI-fueled solutions,” said Jason Girzadas, managing principal, businesses, global, and strategic services (BGS) and CEO Elect, Deloitte US
Although taking advantage of AI, ML and RPA canbenefit an organization, it is also important tounderstand and consider the risk involved:.
1) Using AI tools built by humans introduces the ethics and bias of human judgement and stereotyping.
2) Inadequate testing of AI outcomes can produce questionable results or audit outcomes.
3) Human logic errors.
There are 2 primary aspects auditors should consider while performing the audit of AI applications: ,Compliance —Assess risk related to the rights and freedoms of data subjects
Technology —Assess risk related to machine learning, data science and cybersecurity
Auditing will be a key governance mechanism to confirm that AI systems are designed and deployed in line with a company’s goals
But to create a risk-based audit plan specific to generative AI, Internal Audit must design and adopt new audit methodologies, new forms of supervision and new skill sets
AI is changing every business and organization
In the next decade, Artificial Intelligence (AI) will fundamentally transform the audit through automating procedures and enhancing audit quality
First, the summit participants agreed on three key principles for the use of generative AI in research — accountability, transparency and independent oversight
AccountabilityHumans must remain in the loop to evaluate the quality of generated content; for example, to replicate results and identify bias
