Amid the excitement surrounding agentic AI’s transformative potential, several publications have dubbed 2025 the year of agentic AI. The reality, however, is that the technology remains in its infancy. Despite the blistering pace of innovation, fully autonomous AI agents are, in many respects, still an aspirational goal — held back by significant technical, ethical and regulatory challenges. Take AutoGPT, for instance — the first widely recognised autonomous AI agent that was released in 2023. As an agentic counterpart to ChatGPT (both are powered by OpenAI’s GPT models), AutoGPT has faced persistent criticism for slow processing speeds and glaring inaccuracies, severely limiting its real-world utility. As a result, many still prefer the reliability and responsiveness of ChatGPT. But to fixate on what qualifies as truly “agentic” quickly becomes an exercise in unmitigated pedantry. A more useful lens is to view agentic functionality as existing on a spectrum. Within this framework, the current trend of agentic transformation is less about achieving full autonomy and more about integrating agentic capabilities into existing GenAI workflows — enhancing these processes with improved decision-making, multi-tasking and automation features (see Diagram 1).
The world of artificial intelligence (AI) moves fast, and so does its vocabulary. Today’s buzzword? Agentic AI. Over the past year, Google searches for this term have surged by more than 1,000%. Regular readers may recall that this isn’t the first time we’ve covered the topic here in The Edge. To recap, agentic AI refers to any AI system that has been designed to perform complex tasks with minimal human oversight. While the last big talk of the town — generative AI (GenAI) — revolves around “content”, agentic AI is all about “action” — carried out through AI agents.
In this context, AI agents can be likened to their human counterparts — think insurance or travel “agents” — entities empowered to act on behalf of others (or, in pure economic parlance, their “principals”). Similarly, AI agents operate within defined parameters, making decisions and taking action in accordance with pre-defined goals set by users. Semantically speaking, however, the term “agentic” in agentic AI doesn’t stem from economic agency theory, but from psychology. Specifically, it draws on Albert Bandura’s concept of agentic behaviour: the ability of an entity — human or, increasingly, artificial — to act and pursue goals independently. This degree of autonomy is precisely what distinguishes agentic AI from earlier technologies like Siri or Google Assistant, which primarily functioned as reactive, limited-purpose helpers. Agentic AI marks a paradigm shift — from passive tools to autonomous, goal-driven systems.
