ISSN 2477-1686
Vol. 12 No. 61 Juli 2026
Decoding Intelligence: Should we put our faith in artificial intelligence?
Oleh:
Astrid Gisela Herabadi & Gustav Arkady Kadarusman
Faculty of Psychology, Atma Jaya Catholic University of Indonesia
University of Sheffield, United Kingdom
It is a truth universally acknowledged that nowadays, one can hardly navigate the digital world without coming across some form of artificial intelligence (AI). We can even say that most people have formed the habit of engaging in some form of “communication” with AI. We are starting to rely on AI as a tool to help organise data and facts to improve our knowledge and understanding of the world. Furthermore, the reliance toward AI has progressed into a source of emotional support.
Let’s stop for a moment and ponder this question: How does artificial intelligence compare to humans’ “natural” intelligence? As the name suggests, does AI’s so called “intelligence” equate or even surpass human’s intelligence? And more importantly: how far should we allow them to do the thinking for us?
AI is now developing at an unprecedented speed, to the point of rivalling human performance in certain activities. A recent milestone of AI AlphaGo is winning a game of Go (aka Chinese chequers), which is relatively more complex compared to chess, with magnitudes of different possible moves (Silver et al., 2016). This AI performed several unorthodox moves that human players would not do. These not easily explainable algorithms is a “black box” on AI processes and decision making.
As with any other technology, perception and interaction from human users is an important aspect to consider in AI use. It functions as a technological system with a considerable degree of autonomy and equipped with several human-like capabilities (Krafft et al., 2020). In contrast to other technologies, human users perceive this as agency (Carey & Spelke, 1994) which in turn affects their behaviour when interacting with AI (Cameron et al., 2021). Consequently, it can be implied that trust of AI as an agent; when people believe that they can have an actual “discourse” with AI about every aspect of everyday living; seem to have an impact on attitudes.
The combination of the “black box” processing and the perceived agency of AI by human users seems to be greatly affecting the intention of use; as normally measured through frameworks such as the Theory of Planned Behaviour (TPB) (Ajzen, 1985) and Technology Acceptance Model (TAM) (Davis, 1986). Previous research has integrated trust as a key construct in TAM resulting in a more predictive model (Dickson et al, 2021). A recent study tested an integrated model based on TAM and TPB, with an added factor of trust, as predictors of intention to use AI (Kadarusman, 2024). In allowing for deeper analysis and a holistic understanding on the relationships between the variables, the results of the above-mentioned study were analysed and processed into a Structural Equation Modelling (SEM) (Hair et al., 2013). The final results show that though some factors are generally consistent with the initial hypothesis, some factors show a different or unexpected influence to intention, which was unaccounted for in the initial conceptual model. Most notably, trust has a much more significant effect than expected from other technologies, which to a certain extent has diminished other factors contributing to intention to use. We can conclude from this that humans do generally perceive AI with greater agency than normal, and as such place a great emphasis on perceived trust.
Also important is that AI is increasingly performing and automating tasks that humans currently do, which leads to a perception of human qualities such as reasoning and motivation (Glikson & Woolley, 2020) and may cause misconceptions from human users that AI possess a similar intelligence as normally applied when referring to living things. To emphasise, it can be dangerous to conflate this into a singular concept of intelligence. In principle, AI are intrinsically still thinking machines, albeit with far greater capabilities than in the past.
In short, believing that AI will ALWAYS performs better and provides better results or judgement is in fact a fallacy, because like “natural” intelligence there is variation of how intelligent AI can process a “cognitive” task. According to the multiple cognitive mechanism theory of intelligence, human general cognitive ability, sometimes referred to as the g-factor, likely comprises of multiple interacting paths of processes (Kaufman et al, 2013). Mechanisms of working memory, processing speed, and explicit associative learning are involved when humans are faced with a cognitive task.
Unsurprisingly, most people are awed by AI’s performance since it obviously excels in its ability to maintain, update, and manipulate information in the face of distraction and competing representations (working memory) and the speed at which it performs cognitive operations (processing speed). However, the third mechanism, explicit associative learning, proves to be a greater challenge to AI, since it involves the ability to remember and voluntarily recall specific associations between stimuli. The tacit skills and innate human abilities to draw conclusions based on intuitions and common sense, which is deeply rooted in embodied experience through implicit learning and context, are difficult to mimic (Kuzma, 2025). As a result, when AI is presented with a relatively new or rarely explored task, it fails to gather enough relevant information and makes a “leap-of-faith”, jumping to a wrong conclusion.
References:
Ajzen, I. (1985). From Intentions to Actions: A Theory of Planned Behavior. In J. Kuhl & J. Beckmann (Eds.), Action Control: From Cognition to Behavior (pp. 11–39). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-69746-3_2
Cameron, D., de Saille, S., Collins, E. C., Aitken, J. M., Cheung, H., Chua, A., Loh, E. J., & Law, J. (2021). The effect of social-cognitive recovery strategies on likability, capability and trust in social robots. Computers in Human Behavior, 114, 106561. https://doi.org/10.1016/j.chb.2020.106561
Carey, S., & Spelke, E. (1994). Domain-specific knowledge and conceptual change. In L. A. Hirschfeld & S. A. Gelman (Eds.), Mapping the Mind: Domain Specificity in Cognition and Culture (pp. 169–200). Cambridge University Press. https://doi.org/DOI: 10.1017/CBO9780511752902.008
Davis, F. D. (1986). A technology acceptance model for empirically testing new end-user information systems: Theory and results. Massachusetts Institute of Technology.
Dickson B.U., Oby B.O., Samuel N.N., Udoka S.O. (2021), Integrating Trust into Technology Acceptance Model (TAM), the Conceptual Framework for E-Payment Platform Acceptance. British Journal of Management and Marketing Studies 4(4), 34-56. DOI: 10.52589/BJMMSTB3XTKPI.
Glikson, E., & Woolley, A. W. (2020). Human Trust in Artificial Intelligence: Review of Empirical Research. Academy of Management Annals, 14(2). https://doi.org/https://doi.org/10.5465/annals.2018.0057
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2013). Multivariate data analysis : Pearson new international edition. Pearson Education, Limited.
Kadarusman, G. (2024). Factors Affecting Attitudes and Beliefs Towards Artificial Intelligence [Master’s Thesis]. The University of Sheffield.
Kaufman, J. C., Kaufman, S. B., & Plucker, J. A. (2013). Contemporary theories of intelligence. In D. Reisberg (Ed.), The Oxford handbook of cognitive psychology (pp. 811–822). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780195376746.013.0051
Krafft, P. M., Young, M., Katell, M., Huang, K., & Bugingo, G. (2020). Defining AI in Policy versus Practice. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, AIES ’20, 72–78. https://doi.org/10.1145/3375627.3375835
Kuzma, J.D. The Irreplicable Nature of Human Intuition: A Critical Response to Claims of Artificial Intuition. Philos. Technol. 38, 59 (2025). https://doi.org/10.1007/s13347-025-00894-5
McCarthy, J., Minsky, M. L., Rochester, N., & Shannon, C. E. (2006). A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence.
