arbiter
JournalsPricingAccount

Examining different artificial intelligence models’ ability to pass Certificate of Theory in Accountancy-level tax questions

Asheer Ram & Wayne van Zijl

South African Journal of Economic and Management Sciences2026https://doi.org/10.4102/sajems.v29i1.6348article
AJG 1ABDC C
Weight
0.50

What the paper says

As Artificial Intelligence (AI) models become more sophisticated and entrenched in accountancy professions, this raises questions about their ability to outperform humans. This article is one of the first to examine the ability of five different AI models to pass professional tax examinations. Contribution: This article provides evidence about AI’s current ability to support or replace tax practitioners. It provides a baseline to track the progress of different AI models as they evolve. Only Grok passed, while ChatGPT, Claude, CoPilot, and Gemini failed. Notably, the AI models provided persuasive answers despite being incorrect, negating their ability to replace tax practitioners.

Open paper page →

Cite this paper

https://doi.org/https://doi.org/10.4102/sajems.v29i1.6348

Or copy a formatted citation

@article{asheer2026,
  title        = {{Examining different artificial intelligence models’ ability to pass Certificate of Theory in Accountancy-level tax questions}},
  author       = {Asheer Ram & Wayne van Zijl},
  journal      = {South African Journal of Economic and Management Sciences},
  year         = {2026},
  doi          = {https://doi.org/https://doi.org/10.4102/sajems.v29i1.6348},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Examining different artificial intelligence models’ ability to pass Certificate of Theory in Accountancy-level tax questions

Flags are reviewed by the Arbiter methodology team within 5 business days.

Evidence weight

0.50

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.50 × 0.4 = 0.20
M · momentum0.50 × 0.15 = 0.07
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.