The rapid advancement of artificial intelligence has introduced profound implications for professional fields, creating unique challenges for education and assessment. As tools like ChatGPT demonstrate an ability to answer technical multiple-choice questions with near-perfect accuracy, accounting professionals and educators are forced to re-examine long-standing methods of verifying continuing professional education (CPE).

This intersection of advanced technology and professional requirements places significant pressure on self-study programs, where qualified assessments are traditionally used to validate comprehension and engagement with technical material.

The Regulatory Silence on Artificial Intelligence in CPE

The emergence of generative AI has created severe ethical questions regarding the integrity of self-study CPE programs. Current regulatory frameworks, however, offer little direction. As of the most recent 2024 revision, the Statement on Standards for CPE Programs—issued jointly by the National Association of State Boards of Accountancy (NASBA) and the AICPA—makes no reference to artificial intelligence or its usage in assessments.

This lack of explicit guidance leaves substantial room for interpretation. Because the standards do not explicitly prohibit seeking external help or utilizing AI tools—restricting only the provision of feedback on failed attempts—participants face a gray area. Furthermore, because other formats like Group Live or Group Internet do not require qualified assessments, some argue that these tests serve primarily to promote engagement rather than strict mastery. Consequently, leveraging AI can sometimes be viewed as a novel method of interacting with material or reflecting evolving professional competencies rather than outright cheating.

Rethinking Assessment Design to Reduce Temptation

Despite differing interpretations of compliance, the ethical implications remain serious. Experts suggest that the ethical onus should not rest solely on CPE participants. Instead, providers hold a responsibility to design systems that minimize strong inducements to behave unethically.

Superficial technical fixes, such as disabling the ability to copy and paste text, are easily circumvented. Instead, educators and program designers are exploring more creative approaches to evaluation:

  • Contextualizing Questions: Designing multiple-choice questions that are less self-contained and require specific clues subtly embedded within proprietary learning materials.
  • Leveraging External Nuance: Requiring participants to connect technical changes across different time periods or documentation formats that generic language models cannot easily solve without specific course context.
  • Offering Transparent Dual-Path Systems: Providing participants a clear, transparent choice between taking an assessment without AI under a standard passing threshold, or using AI under a significantly higher passing threshold (such as 95 percent).

Testing alternative approaches—such as dual-path grading models in college online exams—indicates that when individuals are presented with a fair and transparent choice, they are more inclined to engage in good faith while respecting individual autonomy.

Moving Toward Human-Centered Educational Integrity

While generative artificial intelligence models present legitimate threats to the traditional integrity of self-study assessments, relying on restrictive enforcement or simple self-certification may place professionals in unnecessary ethical binds. By acknowledging the shifting landscape of professional competency and adapting evaluation frameworks through thoughtful design, educational providers can better uphold both the letter and the spirit of professional education.