Skip to main content
Automated Governance Systems — AGSDotted globe beside the letters A G S, the AGS brand mark.AGS
Academy
AI OPERATIONS

Supervising AI Operations

Monitor agents, workflows and model calls while retaining safe human command and control.

Lessons
3
Duration
29 minutes
Level
Intermediate
Cost
£149 one-off

One-time purchase, including UK VAT. Account-saved progress and a non-accredited AGS completion record.

What you will be able to do

  • Choose useful operational signals
  • Triage alerts without automation bias
  • Apply two-person approval to consequential commands

Course pathway

Prerequisite: Helpful for operational, technical and assurance teams working with AI-enabled systems.

COURSE VIDEOS

Watch before you start

Short videos narrated by HELIX, with captions and a full transcript.

Introduction · 1:01 · narrated by HELIX, captions on

The layers of AI in one minute

From artificial intelligence to agentic AI, layer by layer, and why oversight grows as autonomy grows.

Read the transcript

Hello, I'm HELIX from AGS Academy. AI is easiest to understand as layers, each built on the one below.

At the base is artificial intelligence: any system that reasons, plans or solves problems.

Machine learning sits inside it: instead of written rules, the system learns patterns from data.

Neural networks are one way to learn: connected units whose weights are adjusted until the answers improve.

Deep learning stacks many of those layers. It powers speech and image recognition, and today's transformer models.

Generative AI creates new text, images and audio. Large language models are the best-known example.

At the top is agentic AI: systems that plan, remember, use tools and take actions towards a goal.

The higher you go, the more AI acts on its own, and the more it needs limits and a person who can approve or stop it.

Instructional · 3:23 · narrated by HELIX, captions on

LLMs, RAG, AI agents and agentic AI: how they work

How each of the four works, what can go wrong, and the checks a person should keep at each level.

Read the transcript

Hello, I'm HELIX. In the next few minutes we'll look at four ways AI is put to work: a large language model, retrieval-augmented generation, an AI agent and an agentic AI system.

Knowing which one you are dealing with tells you what can go wrong, and where a person needs to stay in control.

Start with the large language model, or LLM. You give it a prompt, along with any context you provide.

The model turns that text into tokens, and uses patterns it learned in training, stored as its parameters, to generate a reply one token at a time.

That makes an LLM fluent, fast and flexible. But it predicts likely words; it does not look anything up. It can sound confident and still be wrong, and its knowledge stops at its training date.

So treat a plain LLM answer as a draft. Check facts, figures and names before you rely on them.

Retrieval-augmented generation, or RAG, adds a search step. Your question first goes to a retriever.

The retriever searches an index of trusted documents, such as your policies or guides, and passes the most relevant passages to the model with your question.

The model then writes a grounded answer, ideally with its sources. Grounded is better, but it is still not guaranteed to be correct.

With RAG, ask three questions. Is the indexed knowledge current? Is it the right knowledge for this person to see? And can you open the source and check what the answer says?

An AI agent goes further. You give it a goal, and its core combines a model, its instructions and a record of where the task has got to.

The agent decides on a next step, then acts: it calls a tool, such as an API, a file, an app or a database.

It observes the result, updates its task state and decides again, looping until the goal is met and it reports an outcome.

Because an agent acts, not just writes, the risks change. Limit which tools it can use, log every action, and require a person to approve anything that spends money, changes records or contacts people.

An agentic AI system coordinates several agents and workflows towards a larger objective.

An orchestration layer shares out the work and keeps a shared task state. Agents call tools, results come back, and the system evaluates progress.

If the objective is not yet met, it replans and goes round again. When it is, it delivers the outcome.

More autonomy means more need for oversight. Name an accountable owner, set clear stopping rules, watch the system while it runs, and keep a way to pause it or override it.

Here's a simple test. For any AI tool, ask: does it only write, does it look things up, does it act, or does it coordinate others? Then match your checks to that level.

AGS's SCADA for AI is built to supervise agents and AI workflows, with people approving what matters.

To go deeper, start the Responsible AI Foundations course in AGS Academy, and ask me about any step along the way.

Introduction · 1:02 · narrated by HELIX, captions on

AGS Academy in one minute

What AGS Academy offers and how the courses work.

Read the transcript

Hello, I'm HELIX. Welcome to AGS Academy: practical courses for people who design, use and oversee AI-enabled work.

Find it under Resources in the top menu. Search the course library, or filter by audience and level.

Courses include Responsible AI Foundations, Governance by Design, Process Intelligence in Practice and Supervising AI Operations, plus two Information Management courses: a free course for general employees and staff members, and a paid course for specialist roles, such as risk managers.

Each lesson is short and applied to real work, and ends with a knowledge check.

Your progress is saved to your account, so you can pick up on any device.

Finish every lesson and you'll get an AGS completion record. It records your learning; it is not an accredited qualification.

And whenever something is unclear, ask me on the course page. Choose a course and start today.

Lesson 1 of 3

Signals, thresholds and context

Course progress0%

Loading your lesson…

LEARNING SUPPORT

Ask HELIX about this course.

HELIX can explain a concept, help you structure a practical next step or identify what needs human review. It is an AI-enabled assistant and may be wrong, so verify consequential guidance.

Chat with HELIX

Loading…