Skip to content
The Art of AI
TwitterLIMukesh K

What are AI Agents

— AI Agents, Definition — 3 min read

In traditional AI, agents are defined as entities that perceive and act upon their environment [5]. In the LLM era, the term is used in a narrower way (a thermostat would qualify as an agent under the traditional definition).

Many researchers have tried to formalize the community’s intuitive understanding of what constitutes an agent in the context of language-model-based systems. Many of them view it as a spectrum — sometimes denoted by the term ‘agentic’ [4] — rather than a binary definition of an agent.

Let's identify the factors that cause an AI system to be considered more agentic according to existing definitions. There are three clusters of factors.

Environment and goals.

The more complex the environment — e.g. range of tasks and domains, multi-stakeholder, long time horizon, unexpected changes — the more AI systems operating in that environment are agentic [6, 3]. Systems that pursue complex goals without being instructed on how to pursue the goal are more agentic [6, 1, 3].

User interface and supervision.

AI systems that can be instructed in natural language and act autonomously on the user’s behalf are more agentic [3]. In particular, systems that require less user supervision are more agentic [6, 1, 3].

System design

Systems that use design patterns such as tool use (e.g., web search, programming) or planning (e.g., reflection, subgoal decomposition) are more agentic [7, 4]. Systems whose control flow is driven by an LLM, and hence dynamic, are more agentic [7, 2].

References