Our AI writing assistant, WriteUp, can assist you in easily writing any text. Click here to experience its capabilities.

Auditors find ways to examine AI despite lack of framework

View Original View Raw

Summary

Auditors are currently developing new practices and procedures for Artificial Intelligence (AI) and finding ways to apply existing ones to this technology, despite lacking an official framework or set of standards. Auditors are using existing standards, analogs from other organizations, and frameworks from the Big Four firms to guide their AI audits. Auditors will need to adapt their approach to AI due to the "black box" nature of the technology, and may need to use experts to help with highly complex AI procedures. Regulations such as the European Union's Artificial Intelligence Act and the Canadian Digital Charter will provide further guidance on AI audits and oversight. Accountants are well-suited to provide assurance and oversight for AI.

Q&As

What do auditors need in order to properly examine AI?
Auditors need existing standards, frameworks, and guidance from organizations such as the AICPA, PCAOB, and the International Internal Audit Standards Board in order to properly examine AI.

What existing standards do auditors currently apply to AI?
Auditors currently apply existing standards concerning technology controls to AI.

What frameworks have organizations created to assist with AI auditing?
Organizations have created frameworks such as the Committee of Sponsoring Organizations of the Treadway Commission's framework for enterprise risk management, the National Institute of Standards and Technology's AI risk management framework, the International Standards Organization, the Cloud Security Alliance, and the Big Four firms' own frameworks for responsible AI usage to assist with AI auditing.

How does AI differ from traditional auditing?
AI tends to be more of a "black box," meaning it can be difficult to impossible to really understand what is happening to produce the outputs it does. This means the auditor can't examine the code or understand how the algorithm works, which is different from traditional auditing.

What role do technical specialists and subject matter experts play in AI audits?
Technical specialists and subject matter experts play a role in AI audits by providing the auditor with the information needed to do the audit effectively and by helping to identify where AI or advanced approaches are being used. They also help to audit the technology and make sure the financial auditor is getting the information he needs from a technical standpoint.

AI Comments

👍 It is great to see auditors finding creative ways to audit AI despite the lack of an official framework. The Big Four firms have all released their own frameworks for responsible AI usage, which can be a great guide for auditors.

👎 Despite the efforts of auditors to find creative ways to audit AI, the lack of an official framework still poses a significant challenge. Without an authoritative framework, there is less consistency from audit to audit and entity to entity.

AI Discussion

Me: It's about how auditors are finding ways to examine AI despite the lack of an official framework or set of standards on artificial intelligence. It looks at the different ways auditors are adapting existing procedures and frameworks to AI and how they are using technical specialists to do so. It also examines some of the larger regulatory frameworks in motion that will eventually provide guidance for auditing AI.

Friend: That's really interesting. It sounds like auditors are really having to get creative when it comes to examining AI.

Me: Yeah, it's definitely a challenge. On the one hand, it's great that auditors are finding ways to incorporate AI into existing processes and frameworks, but on the other hand, they're still using analogs that aren't specifically made for these situations. It's also concerning that until a framework is established, people won't really be able to audit AI because there won't be anything to audit against.

Action items

Research the existing frameworks and standards for AI audit, such as the Committee of Sponsoring Organizations of the Treadway Commission's framework for enterprise risk management and the National Institute of Standards and Technology's AI risk management framework.
Develop a basic understanding of AI and the kinds of questions to ask during an AI audit engagement.
Consider upskilling auditors with technical experts in AI to ensure accuracy and reliability of the audit.

Technical terms

AI (Artificial Intelligence): AI is a type of computer technology that is designed to simulate human intelligence and behavior. It is used to create systems that can think, learn, and act on their own.
Auditors: Auditors are professionals who examine financial records and statements to ensure accuracy and compliance with laws and regulations.
Framework: A framework is a set of rules, guidelines, or principles that provide structure and guidance for a particular activity or process.
Technology Controls: Technology controls are measures taken to ensure the security and reliability of computer systems and networks. They can include access control, data encryption, and authentication.
Algorithm: An algorithm is a set of instructions or rules used to solve a problem or accomplish a task.
Outputs: Outputs are the results of a process or system.
Regulatory Environment: The regulatory environment is the set of laws, regulations, and other rules that govern the activities of businesses and other organizations.
Assurance: Assurance is a type of audit that provides independent assurance that a company's financial statements are accurate and reliable.
Provenance: Provenance is the origin or source of something.
Unbiased: Unbiased means free from bias or prejudice.