RA · Foundations
Learning artificial intelligence and machine learning: a guide to the field
A learning guide to artificial intelligence and machine learning: what an AI system is, how models are governed, a suggested study order and recorded learning.
- As of
- 17 November 2026
- 阅读
- 11 minutes
- 证据
- 10 sources, cited in order
- 级别
- Introductory to intermediate
- 学习时长
- 15 minutes
What you will be able to do
- Give a working definition of an AI system and name its lifecycle dimensions.
- Explain why data change and use matter to risk.
- Name the main principles, rules and standards bodies in the field.
- Describe a simple monitoring plan for a deployed model.
- Find the courses of the discipline and read their pages for what is open.
- 级别
- Introductory to intermediate
- 学习时长
- 15 minutes
- Before you start
- The learning guide to the Faculty of Technology and Digital Systems, or equivalent background
01What artificial intelligence and machine learning covers
What artificial intelligence and machine learning covers
Artificial intelligence and machine learning is the Academy discipline of artificial intelligence and machine learning, one of the disciplines of the Faculty of Technology and Digital Systems. On the Academy's own description it covers how models are built from data and training pipelines, how they are deployed in science and public services, and the reliability, security and societal effects that follow.
It sits in the Department of Artificial Intelligence and Autonomous Systems and borders AI safety and evaluation, AI assurance and audit and robotics and autonomous systems. Its subtopics include machine learning for risk professionals, foundation models and generative AI, data and training pipelines, AI failure modes and robustness, agentic systems, AI security, AI in public services, AI in science, and the labour and social effects of AI.
02Why AI needs its own kind of risk thinking
Why AI needs its own kind of risk thinking
AI systems learn patterns from data instead of following rules a person wrote line by line. That makes them useful for tasks that are hard to write down, and it also changes how they fail.
The NIST AI RMF says the risks posed by AI systems are in many ways unique. For example, AI systems may be trained on data that can change over time, sometimes significantly and unexpectedly, affecting functionality and trustworthiness in ways that are hard to understand. It also describes AI systems as inherently socio-technical 1.
The reach of the technology explains the policy attention. The OECD AI Principles page says that the European Union, the Council of Europe, the United States, the United Nations and other jurisdictions use the OECD's definition of an AI system and lifecycle in their legislative and regulatory frameworks and guidance 2.
03A suggested learning order in four stages
A suggested learning order in four stages
The Academy describes study at four levels: introductory, intermediate, advanced and integrative. The four stages below follow the same climb, but they are about subjects, not levels.
Stage one: what an AI system is. Learn the working definition and the idea of a lifecycle. NIST adapts its definition from the OECD Recommendation on AI and from ISO/IEC 22989, and notes that AI systems are designed to operate with varying levels of autonomy 1. The guide to technology and digital systems gives the wider field.
Stage two: how models are built and used. Follow data into a model and the model into a use. NIST lists four lifecycle dimensions: application context, data and input, AI model, and task and output 1. The OECD classification framework helps policy-makers, regulators and others to assess the opportunities and risks that different types of AI systems present 6.
Stage three: risk and governance. Read the principles and the rules. The OECD AI Principles include values-based principles on inclusive growth, human rights and democratic values, transparency and explainability, robustness, security and safety, and accountability 2. The EU AI Act defines four levels of risk for AI systems 3. CEN and CENELEC Joint Technical Committee 21 develops European AI standards 4.
Stage four: general-purpose models and society. Study foundation models and the effects of AI on work and public services. The Bletchley Declaration describes frontier AI as highly capable general-purpose models, including foundation models, that could perform a wide variety of tasks 7. The field of The Global Risks Alliance (GRA) on AI and models and the discipline's courses extend it.
The order is a suggestion. A policy analyst may begin at stage three; a data scientist may begin at stage two. What matters is that each new idea has the ideas beneath it in place.
Fig. 01Four stages of study in artificial intelligence and machine learning
Figure 1 in words
Four steps rise from left to right. Step one, what an AI system is: the working definition and the lifecycle. Step two, how models are built and used: the four lifecycle dimensions and the OECD classification framework. Step three, risk and governance: the OECD AI Principles and the EU AI Act. Step four, highlighted, general-purpose models and society: foundation models and effects on work and public services.
04From a use to a model and back
From a use to a model and back
Figure 2 follows the lifecycle as NIST describes it. It starts with the application context: who will use the system, for what and where. It moves to the data and input that feed the system, then to the model, then to the task and output that people see. Monitoring in use feeds back into the first step.
The figure is useful because it shows where each kind of failure lives. A poor fit to the context is a design failure. Poor data is an input failure. A weak model is a method failure. A misleading output is a presentation failure. NIST adds that test, evaluation, verification and validation matter throughout the lifecycle 1. A learner who can place a failure on this path can choose the right remedy.
Fig. 02From a use to a model and back
Figure 2 in words
Five layers stacked from top to bottom. One, application context: who uses the system, for what and where. Two, data and input. Three, highlighted, the AI model. Four, task and output: what people see and use. Five, monitoring in use, which feeds back to the application context.
05Five core ideas every learner should hold
Five core ideas every learner should hold
Five ideas carry most of the weight in this discipline. Each is simple to state and takes practice to use.
Models learn from data. What they learn can only be as good as what they were shown 1.
Data changes. A model trained on yesterday's data may drift away from today's world 1.
Outputs are predictions. They are estimates, recommendations or decisions, not facts 1.
AI is socio-technical. People, organisations and settings shape results as much as code 1.
Risk depends on use. The same technique can be low-risk in one setting and high-risk in another 3.
06Worked example: a model that drifts
Worked example: a model that drifts
Worked example (teaching values): the numbers in this section are invented for teaching and describe no real system. A team trains a model to predict whether a delivery will be late. On a held-out test of a hundred cases, it gets ninety right.
Worked example, continued (teaching values): six months later, a new route network opens and customer behaviour changes. Checked on a fresh set of a hundred cases, the model gets seventy-five right. Nothing in the code changed. The data it meets changed.
The lesson is that a model's accuracy belongs to a time and a setting. The team should monitor accuracy in use, set a level at which the model is retrained or withdrawn, and record the date of each check. Stage two teaches the lifecycle that makes this possible.
07Canonical references and bodies to know
Canonical references and bodies to know
A learner in this discipline should know a short list of references, what kind of evidence each holds and how often it is renewed. Primary documents come first; summaries are useful only after them.
For definitions and principles: the OECD AI Principles and the OECD classification framework 26. For risk management: the NIST AI RMF 18. For law: the EU AI Act 3. For standards: CEN-CENELEC JTC 21 4. For UN-system resources: AI for Good, organised by the ITU 5. For the state of the science and trend data: the International AI Safety Report and the Stanford AI Index 910. For frontier risk: the Bletchley Declaration 7.
08How to read a trend report
How to read a trend report
A learner meets trend reports early and reads them badly if nobody explains their shape. Three habits help. First, find the issuer and the stated purpose. The Stanford AI Index states a mission to provide unbiased, rigorously vetted and globally sourced data about AI. That is the issuer's own description, so check the method pages too 10.
Second, find the edition. The International AI Safety Report was published in a second edition in February 2026 and describes itself as a comprehensive review of the latest scientific research on the capabilities and risks of general-purpose AI systems 9. Third, ask what is measured. A count of papers, a benchmark score and a survey of firms answer different questions. Write the measure, the date and the source beside any figure you take from a report.
09Skills that make the knowledge usable
Skills that make the knowledge usable
Knowledge becomes useful through a small set of skills. The discipline treats these as its practical core.
Framing a problem. Say what is to be predicted and for whom. Reading data. Find where it came from, what it leaves out and how it may change. Splitting and testing. Keep test data apart from training data. Reading a metric. Know what accuracy hides. Monitoring. Watch results in use and set triggers. Explaining. State what a model does and does not do in plain words. Reading a rule. Find the risk level and the duties that follow.
10How expertise is shown at the Academy
How expertise is shown at the Academy
Expertise in this field is shown by work that others can check. The Academy records learning, not credentials.
Finishing a course gives a completion record. It is not a certificate, an accreditation or a qualification, and it does not state that a person is competent in a profession. Achievements on the Academy recognise assessed learning only.
The course catalogue page for this discipline lists two courses: Artificial intelligence and machine learning: foundations and practice, at the intermediate level, and Artificial intelligence and machine learning: frontier methods, at the advanced level. On 17 November 2026 the page shows the first as in authoring and the second as planned. The discipline page lists one primary course in preparation and four courses taught here from other disciplines. The studios page for technology and digital systems also describes an integrative-level team studio for the department, which takes one real place through its disciplines and ends in a decision brief. The catalogue says the NXS Learning catalogue lists every open course and is where you enrol, so this guide does not point to an Academy course as a place to enrol. Check the catalogue before you plan.
Outside the Academy, the strongest evidence of expertise is a body of work: a documented model with its data and limits stated, a reproducible experiment that cites its sources and dates, or a clear brief that a reviewer has checked. Keep such a record from your first course.
11Emerging areas to watch
Emerging areas to watch
Several areas are changing what an AI practitioner needs to know. Each is a place to read on once the core is in place.
General-purpose models. The EU AI Act puts in place rules for providers of general-purpose AI models, for example on transparency and copyright 3. Standards. JTC 21 aims to develop harmonised standards for legal conformity under the AI Act 4. Gaps in governance. The 2026 AI Index says the frameworks needed to govern, evaluate and understand AI are falling behind the technology 10. Shared resources. AI for Good lists a UN AI Resource Hub among its AI standards and policy pages 5.
Related work across the Nexus network gives wider context. See the field page of The Global Centre for Risk and Innovation (GCRI) on artificial intelligence, the page of The Global Risks Forum (GRF) on AI and digital governance and the proposed Nexus Standards for digital and AI systems. These pages describe programmes and proposals. A link to them implies no endorsement of the sources this guide cites.
What this guide does not claim: it is suggested learning, not a curriculum that anyone must follow, and it is not legal, technical or professional advice. It reports figures as their issuers published them and draws no new estimates from them; the only invented numbers are the labelled teaching values. It names bodies and reports for learning without implying any partnership. Risk Academy does not certify competence, and a completion record is not a professional qualification. The Nexus Standards are proposed and in draft, not adopted. The Academy records no partners, cohorts or learner outcomes for this discipline, and this guide implies none.
常见问题
问与答
Where should a complete beginner start in artificial intelligence and machine learning?
Start with stage one, what an AI system is. Read the NIST working definition and the OECD description of the AI lifecycle, then follow data into a model and a model into a use. Check the course catalogue for the discipline before you plan, because the Academy courses were not yet open for enrolment on 17 November 2026.
What is the difference between artificial intelligence and machine learning?
Artificial intelligence is the wider term for systems that generate outputs such as predictions, recommendations or decisions for given objectives. Machine learning is a way of building such systems by learning patterns from data instead of writing every rule by hand.
Why do AI systems need monitoring after release?
Because the data they meet can change over time, sometimes significantly and unexpectedly. NIST notes that this can affect functionality and trustworthiness in ways that are hard to understand. A model that passed a test once may not pass it later.
Which rules apply to AI?
It depends on place and use. The OECD AI Principles are a shared reference, and the EU AI Act sets risk-based rules with four levels of risk. Many other bodies publish voluntary frameworks. Read the status and scope of each text before you rely on it.
Does finishing a course in this discipline give me a professional qualification?
No. A completion record shows that you finished a course. It is not a certificate or a professional qualification, and the Academy does not certify competence. If you need a recognised qualification, check the requirements of the body that awards it.
条款
术语表
- AI system
- A machine-based system that generates predictions, recommendations or decisions for given objectives.
- Machine learning
- Building a model by learning patterns from data.
- 基础模型
- A general-purpose AI model that can perform a wide variety of tasks.
- Lifecycle
- The stages from context and data to model, output and monitoring.
- 漂移
- A change in the data a model meets that reduces its accuracy over time.
- Socio-technical
- Shaped by people, organisations and settings as well as by technology.
相关内容
Further reading across the network
- 人工智能与机器学习
- Courses of the discipline
- 技术与数字系统学院
- 人工智能与自主系统系
- Learning guide: technology and digital systems
- Learning guide: AI safety and evaluation
- Learning guide: AI assurance and audit
- Learning guide: cybersecurity and digital resilience
- GRA: AI and models
- GCRI: Artificial intelligence
- GRF: AI and digital governance
- Nexus Standards (proposed): digital and AI
Check your understanding
Name the four lifecycle dimensions in the NIST framework.
Application context, data and input, AI model, and task and output.
Why can a model's accuracy fall without any change to its code?
Because the data it meets can change. Accuracy belongs to a time and a setting.
Is the OECD AI Principles page a legal text?
It describes principles and recommendations adopted in 2019 and updated in May 2024. Read the status of each text before you rely on it.
Continue with a course
- Artificial intelligence and machine learning: foundations and practice artificial-intelligence.core
- Artificial intelligence and machine learning: frontier methods artificial-intelligence.frontier
- Artificial Intelligence and Autonomous Systems studio studios-technology-digital.ai-autonomous-systems
来源
参考文献
10 sources, numbered in the order they are first cited
- U.S. National Institute of Standards and Technology (2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1. nvlpubs.nist.gov. Accessed 2026-11-17.
- OECD.AI Policy Observatory (2026). OECD AI Principles overview. oecd.ai. Accessed 2026-11-17.
- European Commission (2026). AI Act: regulatory framework for AI. digital-strategy.ec.europa.eu. Accessed 2026-11-17.
- CEN and CENELEC (2026). Artificial Intelligence: CEN and CENELEC Joint Technical Committee 21. www.cencenelec.eu. Accessed 2026-11-17.
- International Telecommunication Union (2026). About us: AI for Good. aiforgood.itu.int. Accessed 2026-11-17.
- OECD.AI Policy Observatory (2026). OECD Framework for the Classification of AI Systems. oecd.ai. Accessed 2026-11-17.
- UK Government (2023). The Bletchley Declaration by countries attending the AI Safety Summit. www.gov.uk. Accessed 2026-11-17.
- U.S. National Institute of Standards and Technology (2026). AI Risk Management Framework. www.nist.gov. Accessed 2026-11-17.
- International AI Safety Report (2026). International AI Safety Report. internationalaisafetyreport.org. Accessed 2026-11-17.
- Stanford Institute for Human-Centered Artificial Intelligence (2026). AI Index. hai.stanford.edu. Accessed 2026-11-17.
记录
Editorial record
Facts and course states are given as of 17 November 2026. This guide is reviewed at least every 270 days and whenever a cited source publishes a new edition or a course of the faculty opens.
Published by the Academy desk of Risk Academy. To report an error or request a correction, write through the Risk Academy contact page; corrections are dated and listed with the page. The Academy's statement on AI use explains how AI assistance is used and reviewed. Text, visuals and data are licensed CC BY 4.0, except where a cited source states other terms.
Cite this article
Academy desk, Risk Academy (2026, November 17). Artificial intelligence and machine learning guide. Risk Academy. https://risksacademy.com/technology-digital/learning-artificial-intelligence-and-machine-learning-guide/
@online{academy2026learningartificialinte,
author = {{Academy desk, Risk Academy}},
title = {{Artificial intelligence and machine learning guide}},
publisher = {Risk Academy},
year = {2026},
month = {11},
url = {https://risksacademy.com/technology-digital/learning-artificial-intelligence-and-machine-learning-guide/},
language = {en}
} TY - ELEC AU - Academy desk, Risk Academy TI - Artificial intelligence and machine learning guide PB - Risk Academy PY - 2026 DA - 2026/11/17 UR - https://risksacademy.com/technology-digital/learning-artificial-intelligence-and-machine-learning-guide/ LA - en ER -
Republish
Republish or translate this article free of charge, in print or online, under CC BY 4.0: keep the byline and Risk Academy, link to the original and note any change. Figures, maps and twins made by the institutions carry the same licence; third-party data keeps its own credit line.
Attribution to paste
<p>Learning artificial intelligence and machine learning: a guide to the field — Academy desk, Risk Academy, Risk Academy. First published 17 November 2026 at <a href="https://risksacademy.com/technology-digital/learning-artificial-intelligence-and-machine-learning-guide/">https://risksacademy.com/technology-digital/learning-artificial-intelligence-and-machine-learning-guide/</a>. Republished under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).</p>