Inartificial intelligence, anintelligent agent is an entity thatperceives its environment, takes actions autonomously to achieve goals, and may improve its performance throughmachine learning or by acquiringknowledge. AI textbooks[which?] define artificial intelligence as the "study and design of intelligent agents," emphasizing that goal-directed behavior is central to intelligence.
A specialized subset of intelligent agents,agentic AI (also known as an AI agent or simply agent), expands this concept by proactively pursuing goals, making decisions, and taking actions over extended periods.
Intelligent agents can range from simple to highly complex. A basicthermostat orcontrol system is considered an intelligent agent, as is ahuman being, or any other system that meets the same criteria—such as afirm, astate, or abiome.[1]
Intelligent agents operate based on an objective function, which encapsulates their goals. They are designed to create and execute plans that maximize the expected value of this function upon completion.[2] For example, areinforcement learning agent has a reward function, which allows programmers to shape its desired behavior.[3] Similarly, anevolutionary algorithm's behavior is guided by a fitness function.[4]
Intelligent agents in artificial intelligence are closely related toagents ineconomics, and versions of the intelligent agentparadigm are studied incognitive science,ethics, and the philosophy ofpractical reason, as well as in manyinterdisciplinarysocio-cognitivemodeling and computersocial simulations.
Intelligent agents are often described schematically as abstract functional systems similar to computer programs. To distinguish theoretical models from real-world implementations, abstract descriptions of intelligent agents are called abstract intelligent agents. Intelligent agents are also closely related tosoftware agents—autonomous computer programs that carry out tasks on behalf of users. They are also referred to using a term borrowed fromeconomics: a "rational agent".[1]
 | This sectionpossibly containsoriginal research. Relevant discussion may be found onTalk:Intelligent agent. Pleaseimprove it byverifying the claims made and addinginline citations. Statements consisting only of original research should be removed.(February 2023) (Learn how and when to remove this message) |
The concept of intelligent agents provides a foundational lens through which to define and understand artificial intelligence. For instance, the influential textbookArtificial Intelligence: A Modern Approach (Russell & Norvig) describes:
Other researchers and definitions build upon this foundation. Padgham & Winikoff emphasize that intelligent agents should react to changes in their environment in a timely way, proactively pursue goals, and be flexible and robust (able to handle unexpected situations). Some also suggest that ideal agents should be "rational" in the economic sense (making optimal choices) and capable of complex reasoning, like having beliefs, desires, and intentions (BDI model). Kaplan and Haenlein offer a similar definition, focusing on a system's ability to understand external data, learn from that data, and use what is learned to achieve goals through flexible adaptation.
Defining AI in terms of intelligent agents offers several key advantages:
Anobjective function (orgoal function) specifies the goals of an intelligent agent. An agent is deemed more intelligent if it consistently selects actions that yield outcomes better aligned with its objective function. In effect, the objective function serves as a measure of success.
The objective function may be:
The objective function encapsulatesall of the goals the agent is designed to achieve. For rational agents, it also incorporates the trade-offs between potentially conflicting goals. For instance, a self-driving car's objective function might balance factors such as safety, speed, and passenger comfort.
Different terms are used to describe this concept, depending on the context. These include:
Goals, and therefore the objective function, can be:
Some AI systems, such asnearest-neighbor, reason by analogy rather than being explicitly goal-driven. However, even these systems can have goals implicitly defined within their training data.[6] Such systems can still bebenchmarked by framing the non-goal system as one whose "goal" is to accomplish its narrow classification task.[7]
Systems not traditionally considered agents, likeknowledge-representation systems, are sometimes included in theparadigm byframing them as agents with a goal of, for example, answering questions accurately. Here, the concept of an "action" is extended to encompass the "act" of providing an answer. As a further extension, mimicry-driven systems can be framed as agents optimizing a "goal function" based on how closely the IA mimics the desired behavior.[2] Ingenerative adversarial networks (GANs) of the 2010s, an "encoder"/"generator" component attempts to mimic and improvise human text composition. The generator tries to maximize a function representing how well it can fool an antagonistic "predictor"/"discriminator" component.[8]
Whilesymbolic AI systems often use an explicit goal function, the paradigm also applies toneural networks andevolutionary computing.Reinforcement learning can generate intelligent agents that appear to act in ways intended to maximize a "reward function".[9] Sometimes, instead of setting the reward function directly equal to the desired benchmark evaluation function, machine learning programmers use reward shaping to initially give the machine rewards for incremental progress.[10]Yann LeCun stated in 2018, "Most of the learning algorithms that people have come up with essentially consist of minimizing some objective function."[11]AlphaZero chess had a simple objective function: +1 point for each win, and -1 point for each loss. A self-driving car's objective function would be more complex.[12] Evolutionary computing can evolve intelligent agents that appear to act in ways intended to maximize a "fitness function" influencing how many descendants each agent is allowed to leave.[4]
The mathematical formalism ofAIXI was proposed as a maximally intelligent agent in this paradigm.[13] However, AIXI isuncomputable. In the real world, an IA is constrained by finite time and hardware resources, and scientists compete to produce algorithms that achieve progressively higher scores on benchmark tests with existing hardware.[14]
An intelligent agent's behavior can be described mathematically by anagent function. This function determines what the agentdoes based on what it hasseen.
Apercept refers to the agent's sensory inputs at a single point in time. For example, a self-driving car's percepts might include camera images, lidar data, GPS coordinates, and speed readings at a specific instant. The agent uses these percepts, and potentially its history of percepts, to decide on its next action (e.g., accelerate, brake, turn).
The agent function, often denoted asf, maps the agent's entire history of percepts to anaction.[15]
Mathematically, this can be represented as
where:
It's crucial to distinguish between theagent function (an abstract mathematical concept) and theagent program (the concrete implementation of that function).
The agent function can incorporate a wide range of decision-making approaches, including:[16]
Russell & Norvig (2003) group agents into five classes based on their degree of perceived intelligence and capability:[17]
Simple reflex agents act only on the basis of the currentpercept, ignoring the rest of the percept history. The agent function is based on thecondition-action rule: "if condition, then action".
This agent function only succeeds when the environment is fully observable. Some reflex agents can also contain information on their current state which allows them to disregard conditions whose actuators are already triggered.
Infinite loops are often unavoidable for simple reflex agents operating in partially observable environments. If the agent can randomize its actions, it may be possible to escape from infinite loops.
A homethermostat, which turns on or off when the temperature drops below a certain point, is an example of a simple reflex agent.[18][19]
A model-based agent can handle partially observable environments. Its current state is stored inside the agent, maintaining a structure that describes the part of the world which cannot be seen. This knowledge about "how the world works" is referred to as a model of the world, hence the name "model-based agent".
A model-based reflex agent should maintain some sort ofinternal model that depends on the percept history and thereby reflects at least some of the unobserved aspects of the current state. Percept history and impact of action on the environment can be determined by using the internal model. It then chooses an action in the same way as reflex agent.
An agent may also use models to describe and predict the behaviors of other agents in the environment.[20]
Goal-based agents further expand on the capabilities of the model-based agents, by using "goal" information. Goal information describes situations that are desirable. This provides the agent a way to choose among multiple possibilities, selecting the one which reaches a goal state. Search andplanning are the subfields of artificial intelligence devoted to finding action sequences that achieve the agent's goals.
ChatGPT and theRoomba vacuum are examples of goal-based agents.[21]
Goal-based agents only distinguish between goal states and non-goal states. It is also possible to define a measure of how desirable a particular state is. This measure can be obtained through the use of autility function which maps a state to a measure of the utility of the state. A more general performance measure should allow acomparison of different world states according to how well they satisfied the agent's goals. The term utility can be used to describe how "happy" the agent is.
A rational utility-based agent chooses the action that maximizes the expected utility of the action outcomes - that is, what the agent expects to derive, on average, given the probabilities and utilities of each outcome. A utility-based agent has to model and keep track of its environment, tasks that have involved a great deal of research on perception, representation, reasoning, and learning.
Learning lets agents begin in unknown environments and gradually surpass the bounds of their initial knowledge. A key distinction in such agents is the separation between a "learning element," responsible for improving performance, and a "performance element," responsible for choosing external actions.
The learning element gathers feedback from a "critic" to assess the agent's performance and decides how the performance element—also called the "actor"—can be adjusted to yield better outcomes. The performance element, once considered the entire agent, interprets percepts and takes actions.
The final component, the "problem generator," suggests new and informative experiences that encourage exploration and further improvement.
According toWeiss (2013), agents can be categorized into four classes:
In 2013, Alexander Wissner-Gross published a theory exploring the relationship betweenFreedom andIntelligence in intelligent agents.[22][23]
Intelligent agents can be organized hierarchically into multiple "sub-agents." These sub-agents handle lower-level functions, and together with the main agent, they form a complete system capable of executing complex tasks and achieving challenging goals.
Typically, an agent is structured by dividing it into sensors and actuators. The perception system gathers input from the environment via the sensors and feeds this information to a central controller, which then issues commands to the actuators. Often, a multilayered hierarchy of controllers is necessary to balance the rapid responses required for low-level tasks with the more deliberative reasoning needed for high-level objectives.[24]
"Intelligent agent" is also often used as a vague term, sometimes synonymous with "virtual personal assistant".[25] Some 20th-century definitions characterize an agent as a program that aids a user or that acts on behalf of a user.[26] These examples are known assoftware agents, and sometimes an "intelligent software agent" (that is, a software agent with intelligence) is referred to as an "intelligent agent".
According toNikola Kasabov in 1998, IA systems should exhibit the following characteristics:[27]
In the context ofgenerative artificial intelligence,AI agents (also referred to ascompound AI systems) are a class of intelligent agents distinguished by their ability to operate autonomously in complex environments. Agentic AI tools prioritize decision-making over content creation and do not require human prompts or continuous oversight.[28]
They possess several key attributes, including complex goal structures, natural language interfaces, the capacity to act independently of user supervision, and the integration of software tools or planning systems. Their control flow is frequently driven bylarge language models (LLMs).[29] Agents also include memory systems for remembering previous user-agent interactions andorchestration software for organizing agent components.[30]
Researchers and commentators have noted that AI agents do not have a standard definition.[29][31][32][33] The concept of agentic AI has been compared to the fictional characterJ.A.R.V.I.S..[34]
A common application of AI agents is the automation of tasks—for example, booking travel plans based on a user'sprompted request.[35][36] Prominent examples includeDevin AI,AutoGPT, andSIMA.[37] Further examples of agents released since 2025 includeOpenAI Operator,[38]ChatGPT Deep Research,[39]Manus,[40] Quark (based onQwen),[41]AutoGLM Rumination,[41] and Coze (byByteDance).[41] Frameworks for building AI agents includeLangChain,[42] as well as tools such as CAMEL,[43][44] Microsoft AutoGen,[45] and OpenAI Swarm.[46]
 | This section maylendundue weight to certain ideas, incidents, or controversies. Pleasehelp improve it by rewriting it tocreate a more balanced presentation. Discuss andresolve this issue before removing this message.(September 2023) |
The concept of agent-based modeling for self-driving cars was discussed as early as 2003.[47]
Hallerbach et al. explored the use of agent-based approaches for developing and validating automated driving systems. Their method involved a digital twin of the vehicle under test and microscopic traffic simulations using independent agents.[48]
Waymo developed a multi-agent simulation environment called Carcraft, to test algorithms forself-driving cars.[49][50] This system simulates interactions between human drivers, pedestrians, and automated vehicles. Artificial agents replicate human behavior using real-world data.
Salesforce's Agentforce is an agentic AI platform that allows for the building of autonomous agents to perform tasks.[51][52]
TheTransport Security Administration is integrating agentic AI into new technologies, including machines to authenticate passenger identities using biometrics and photos, and also for incident response.[53]
