Autonomous systems

Automation, autonomous systems, and agents

The three terms describe different degrees of choice, context, and delegation; they are not interchangeable.

Back to track

Objectives

Compare an automated rule, an autonomous system, and an agent.
Identify the decision that is actually delegated.

Duration: 9 min
Level: Foundational

Prerequisites

No technical prerequisites

Concepts

Deterministic automation
A predetermined rule triggers an intended operation when specified conditions are met.
Autonomous system
The system selects or adapts actions within a bounded context, under explicit objectives and policies; autonomous never means independent of every rule.
Software agent or AI agent
An agent observes inputs, selects a step, and may use a tool; not every AI is an agent, and the word proves no permission, authority, or reliability.

Visual guide

Comparing delegation

Identify the inputs and trigger.
Name the permitted choices and tools.
Check policy, permissions, and authority.
Identify the consequences and expected observation.

Synthetic example

A rule sorts a message; an agent then proposes an action, but cannot execute it without a separate authorization decision.

More choice does not mean more authority.

Lesson scope

This lesson provides vocabulary for describing the degree of delegation and the limits to expect.

What this lesson does not demonstrate

It does not classify or assess a real system, and it proves no capability, security property, or compliance.

Check question

Which description correctly separates an agent from its authority?

An agent may act as soon as it has expressed an intent.

An agent may propose or select a step within boundaries, without this proving its authorization.

Every automated rule is an autonomous agent.

Choose an answer to read its feedback.

Guided local activity

Put the lifecycle in order

Select the stages in the cautious order of a synthetic action lifecycle.

This sequence triggers no action and produces no evidence about a real system.

Local checklist

Delegation map

Describe a fictional scenario without giving it more authority than it has.

Do not use any real-world data, key, account, or instruction.

Sources and limitations

Source repository: https://github.com/SwissTokint/swisstokint-website
Source reference: codex/learning-hub-v3-autonomous-controls
Content commit: 29c88f79cee07a72804bc017b92fab9200fd3734
Source version: learning-v3

Artificial Intelligence Risk Management Framework (AI RMF 1.0)NIST-AIRMF-1.0
Publisher
National Institute of Standards and Technology (NIST)
Version or date
AI RMF 1.0, 2023-01-26
Link verified on
2026-08-13
Scope used
Governance and risk-management framework for AI systems.
Limitation
A voluntary framework proves neither implementation nor conformity.
Agentic AI — Threats and MitigationsOWASP-AGENTIC-2025
Publisher
OWASP Agentic Security Initiative
Version or date
2025-02-17
Link verified on
2026-08-13
Scope used
Emerging threats related to agents and tool use.
Limitation
An emerging guide that is neither exhaustive nor normative.