The swift advancement of AI tools such as ChatGPT, Gemini, and Claude has ignited significant debate within learning facilities regarding their appropriate function in reviews in higher education. Although these tools present innovative opportunities for enhancing learning, creativity, and the exploration of knowledge, they are frequently misused as replacements for traditional search engines. Unlike established platforms like Google or Bing, which continuously update by indexing real-time web content, AI systems generate responses based on pre-existing, static training data. Thus, the data produced may be outdated, unverifiable, and lacking in clear source attribution, raising important concerns around reliability and clarity. In scholarly settings, researchers typically rely on search engines and scholarly databases that provide direct access to abstracts, references, and peer-reviewed literature. Conversely, AI-generated content may mimic scholarly writing but often lacks citations or includes unverifiable claims. A particularly troubling issue is known as "AI hallucination," where AI tools may invent data or references, giving the illusion of authority without factual grounding. Because of these issues, this paper emphasises that AI should not be viewed as a credible or primary source for academic evidence. Rather, such tools should serve a supportive role, helping users generate ideas, summarise complex texts, or clarify difficult concepts. To ensure responsible integration of AI in scholarly applications, academic facilities must create clear protocols and offer training that enables critical analysis of AI outputs, validation of information, and ethical usage. This article aims to explore the authenticity of AI as a citation tool and its limitations in higher education. In conclusion, this discussion promotes a careful, informed utilisation of artificial intelligence in scholarly contexts. With proper surveillance and a clear understanding of its limitations, AI can complement traditional investigative methods while preserving rigour in addition to the trustworthiness of academic scholarship.
Annerose Wanjiku Wang’ang’a (Wed,) studied this question.