An AI agent names a system that pursues a goal across multiple steps with some autonomy: it perceives state, chooses actions, acts through tools or effectors, and reads the results back into its next choice.
Russell and Norvig’s textbook formulation (Artificial Intelligence: A Modern Approach) casts an agent as anything that perceives its environment through sensors and acts on that environment through actuators; the design question becomes how well the mapping from percept history to action serves a performance measure. In current practice the perceive-act core most often wraps a large language model: the model’s output drives tools (a shell, a filesystem, a browser, an API), each tool result conditions the next decision, and because the steps chain, errors compound and recovery matters as much as planning.
The classical taxonomy, in ascending capability:
- simple reflex agents
- model-based reflex agents
- goal-based agents
- utility-based agents
- learning agents
A human hands an agent a goal rather than a procedure, so the engineering discipline around agents centers on delegation: scoped permissions, checkpoints, and a legible record of what the agent actually did.
See also
In the Field’s Own Words
Michael Wooldridge and Nicholas R. Jennings, “Intelligent Agents: Theory and Practice,” The Knowledge Engineering Review 10(2), 1995, §1.1 “A Weak Notion of Agency”:
Perhaps the most general way in which the term agent is used is to denote a hardware or (more usually) software-based computer system that enjoys the following properties: autonomy: agents operate without the direct intervention of humans or others, and have some kind of control over their actions and internal state; social ability: agents interact with other agents (and possibly humans) via some kind of agent-communication language; reactivity: agents perceive their environment, and respond in a timely fashion to changes that occur in it; pro-activeness: agents do not simply act in response to their environment, they are able to exhibit goal-directed behaviour by taking the initiative.
Source: https://www.cs.ox.ac.uk/people/michael.wooldridge/pubs/ker95/subsubsectionstar3_1_1_1.html
X
BLUESKY
MASTODON
LINKEDIN