By Dr Noor Munirah Isa and Dr Norulhidayah Isa

Artificial intelligence (AI) is rapidly becoming a staple of daily life, influencing how people shop, navigate traffic, and interact with digital systems. In Malaysia, the National AI Office (NAIO) launch under the MyDIGITAL initiative marks a significant step toward integrating AI into key sectors to drive innovation and improve public services. However, addressing widespread misconceptions about AI is crucial for these initiatives to succeed. As noted in Mission AI: The New System Technology by Sheikh, Prins, and Schrijvers (2023), demystification is an essential step in embedding AI into society, ensuring that people understand its capabilities, limitations, and role.

Despite its growing presence, AI remains misunderstood. It is either celebrated as a miraculous tool capable of solving humanity's greatest challenges or feared as a force that will replace jobs and compromise privacy. Much of this misunderstanding arises from science fiction portrayals and advanced discussions about Artificial General Intelligence (AGI) and superhuman AI, overshadowing the fundamental essence of how AI is created and operates. Fictional representations in movies like The Terminator or Ex Machina have long shaped public perceptions, presenting AI as self-aware and potentially dangerous. While entertaining, these depictions are far removed from reality. Current AI operates strictly within the boundaries of its programming and lacks the consciousness or intent often attributed to it in fiction.

The gap between perception and reality widens with concepts like Artificial General Intelligence, a theoretical form of AI that matches human intelligence across all domains. Unlike narrow AI, which excels at specific tasks such as language translation or traffic optimization, AGI could learn and solve problems broadly without retraining for new tasks. An example often cited to illustrate AGI is a single AI system that could perform diverse tasks such as diagnosing illnesses, composing music, driving vehicles, and conducting scientific research without requiring task-specific adjustments. While AGI has captured imaginations, it remains an aspirational concept requiring breakthroughs beyond technological capabilities. Mustafa Suleyman, CEO of Microsoft AI, highlights a compelling perspective on AGI, describing it as the "digital companion that every human needs." He envisions AGI as a tool to enhance human capabilities, helping individuals make better decisions and feel more supported and intelligent. This vision emphasizes AGI's potential as a complement to human intelligence rather than a rival, shifting the narrative from fear to collaboration.

Similarly, superhuman AI, which outperforms humans in specific domains, adds another layer of fascination and apprehension. Examples like AlphaZero for chess and AlphaFold for protein folding demonstrate superhuman capabilities in narrow fields. However, the idea of a general superhuman AI surpassing human intelligence in all areas remains speculative, further fuelling confusion about AI's capabilities.

Contrary to dramatic portrayals, most AI systems today are practical tools designed for specific tasks. They seamlessly integrate into everyday applications, often operating invisibly. For instance, navigation apps use real-time traffic data to suggest the fastest routes, online shopping platforms recommend products based on browsing habits, and chatbots handle customer queries with predefined responses. These examples highlight how AI supports daily activities without resembling the autonomous machines of fiction.

Demystifying AI begins with understanding its current role in society. AI is not a self-aware entity, but a tool programmed to process data, recognize patterns, and assist with specific tasks. Clarifying how AI works in practical contexts dispels the notion of AI as either a magical solution to all problems or a looming existential threat. Public confidence in AI also depends on recognizing its limitations. AI cannot think independently or possess human qualities like creativity, empathy, or ethical reasoning. Instead, it performs tasks based on human-created data and algorithms. This understanding is key to seeing AI as a tool to complement human efforts rather than something to fear.

Building trust in AI requires more than dispelling myths; it involves fostering meaningful engagement with the technology. Education is central to this process. By understanding how AI works, questioning its capabilities, and consulting credible sources, individuals can shift their perceptions from fear and speculation to informed acceptance. Open communication is equally crucial. Public forums, workshops, and accessible resources can bridge the gap between complex AI concepts and everyday understanding. These platforms clarify misconceptions and ensure AI's development aligns with societal values and priorities.

As Malaysia continues integrating AI into various aspects of daily life, demystification will play a vital role in ensuring this transformative technology is embraced with confidence and trust. AI can become a trusted partner in shaping a better future for all by focusing on education, transparency, and engagement.

Dr Noor Munirah Isa, Senior lecturer, Department of Science and Technology Studies, Faculty of Science, Universiti Malaya. Dr Norulhidayah Isa, Senior lecturer, College of Computing, Informatics and Mathematics, Universiti Teknologi MARA Cawangan Terengganu.