The concept of "AI agents " has evolved over decades, and there is no single inventor credited with its creation.1 Instead, the idea emerged from a progression of foundational AI research.

However, we can point to key figures and milestones that established the core principles of an AI agent:

1. Foundational Concepts (The "Father of AI")

Alan Turing (1950): While not inventing the agent itself, his seminal paper, "Computing Machinery and Intelligence," and the proposal of the Turing Test, laid the philosophical and conceptual foundation.3 An "agent" is essentially a machine that can exhibit intelligent behavior, which is what Turing was proposing to test.
John McCarthy (1956): Coined the term "Artificial Intelligence" and was a leader at the Dartmouth Conference, the formal birthplace of the field.5 His work and that of his colleagues established the goal of creating intelligent, problem-solving programs.

2. The First Agent-like Systems

Joseph Weizenbaum (1966): Developed ELIZA, one of the earliest conversational AI programs (a type of software agent).6 It simulated a therapist using simple pattern matching and rule-based scripts, marking a first step toward human-computer interaction and the idea of a system acting on behalf of a user.
The Developers of Early Expert Systems (1970s): Systems like DENDRAL and MYCIN were AI agents designed to mimic the decision-making of a human expert in a narrow field (e.g., diagnosing diseases or analyzing molecular structures).8 They were some of the first practical AI systems that could autonomously reason and act within a defined domain.

3. Formalizing the "Agent" Paradigm

The modern, formal definition of an intelligent agent (an entity that perceives its environment and acts to maximize its chance of success) is heavily indebted to academic work in the late 1980s and 1990s.10

The influential textbook Artificial Intelligence: A Modern Approach, authored by Stuart Russell and Peter Norvig, popularized and formalized the concept of the Rational Agent as the central paradigm of AI.11 This framework solidified the idea of an agent as a system that continuously perceives, reasons, and acts.

Also Read: How AI Agent Works

Summary of Evolution

Era	Key Contribution	Example
1950s	Conceptual Foundation (What is machine intelligence?)	Alan Turing's work
1960s	First Conversational Program (Simple Rule-based Agent)	ELIZA (Joseph Weizenbaum)
1970s-80s	Autonomous Reasoning in a Domain (Expert Agents)	MYCIN
1990s	Formalizing the Intelligent Agent Paradigm	Russell & Norvig's textbook
2020s	Modern Autonomous Agents (LLM-based)	Devin, AutoGPT (uses Generative AI)