Skip to content

Artificial Intelligence, Data Analytics and Cyber Security - Laws and Practice · Artificial Intelligence - Introduction and Basics

Key AI Technologies and Applications Explained

Updated 11 October 2026 · Fact-checked

Key AI technologies are natural language processing (understands text and speech), computer vision (reads images and video), robotics (acts in the physical world), generative AI (creates new content) and expert systems (rule-based advice). To answer, name the technology, explain how it works, then link it to a business, finance or governance use.

Understand Key AI Technologies and Applications

Artificial intelligence (AI) is software that performs tasks which normally need human judgement, such as reading, seeing, deciding or writing. Different AI technologies handle different kinds of input. The exam asks you to tell them apart and match each to a real use.

Natural language processing (NLP) lets machines work with human language, written or spoken. It handles tasks such as classifying text, extracting key terms, translating, summarising and powering chatbots. A company can use NLP to scan thousands of contracts, board minutes or customer complaints in minutes.

Computer vision lets machines interpret images and video. Typical tasks are reading text from scanned documents (optical character recognition), matching faces, spotting defects on a production line and checking identity documents during customer onboarding.

Robotics combines machines, sensors and software to act in the physical world, for example warehouse robots or factory arms. Do not confuse it with robotic process automation (RPA). RPA is software that copies repetitive screen tasks such as data entry. It has no physical body and is usually rule-based.

Generative AI creates new content (text, images, code, audio) after learning patterns from very large data sets. Large language models are a common example. Its outputs can sound confident and still be wrong, which is called hallucination. This creates legal risks: confidentiality, copyright, bias and accuracy.

Expert systems are older, rule-based AI. They hold a knowledge base of expert rules and an inference engine that applies the rules to the facts of a case, using IF-THEN logic. They can explain their reasoning, but they do not learn on their own and only work within the rules they were given.

Applications: in business, demand forecasting, chatbots and quality inspection. In finance, credit scoring, fraud detection, KYC checks, algorithmic trading and robo-advice. In governance and compliance, monitoring regulatory changes, screening related-party transactions, reviewing contracts and drafting first versions of minutes or notices, always with human review.

Key rules to remember

NLP
Text or speech → language understanding → classification, extraction, summary or reply
Input is language. Use for chatbots, contract review and sentiment analysis.
Computer vision
Image or video → pattern recognition → identification, reading or detection
Input is visual. Use for OCR, face match and defect detection.
Expert system
Knowledge base (rules) + inference engine + facts → advice
Rule-based and explainable. It does not learn from data by itself.
Generative AI
Learned patterns from training data + prompt → new content
Creates content. Output needs human verification because of hallucination.
Robotics vs RPA
Robotics = physical machine; RPA = software bot following rules
A common trap. RPA is not a physical robot.

How to solve Key AI Technologies and Applications questions

Use this method for any question on AI technologies or applications, whether it asks you to explain, compare or advise on a case.

  1. 1Read the facts and identify the input: text, speech, images, physical movement, rules or content creation.
  2. 2Match the input to the technology: text to NLP, images to computer vision, physical action to robotics, new content to generative AI, fixed rules to an expert system.
  3. 3Define the technology in one or two lines and say how it works in simple terms.
  4. 4Link it to the exact use in the facts, in business, finance or governance.
  5. 5State the benefit, such as speed, accuracy, cost or coverage.
  6. 6State the risk: bias, errors, hallucination, privacy, security, accountability.
  7. 7Conclude with a practical recommendation, such as human review, data protection safeguards and board oversight.

Quickest way: Input-to-technology match

When to use it: Use when you have little time or the question lists several uses and asks you to identify the technology.

  1. Underline the input in the question.
  2. Write the matching technology next to it.
  3. Add one benefit and one risk for each.
  4. Close with one line on human oversight.

Common mistakes in Key AI Technologies and Applications

  • Treating RPA as robotics or as machine learning.

    The word 'robot' in the name misleads students.

    Fix: Say RPA is rule-based software that automates repetitive digital tasks. Robotics involves a physical machine.

  • Saying expert systems learn from data.

    Students mix them with machine learning.

    Fix: Expert systems apply rules written by human experts. Learning from data belongs to machine learning.

  • Calling generative AI output reliable.

    The text looks fluent and professional.

    Fix: Mention hallucination and say a qualified person must verify the output before use.

  • Listing applications without naming the technology.

    Students memorise use cases but not the link to the technology.

    Fix: Write each use as technology plus task, for example 'NLP for reviewing contracts'.

  • Giving only benefits and ignoring risks and governance.

    Students treat the topic as descriptive.

    Fix: End every answer with risks and safeguards: data privacy, bias, accountability and human oversight.

Worked examples

Example 1

A listed company wants to reduce time spent reviewing vendor contracts and verifying scanned KYC documents of new customers. Identify the AI technologies suitable for each task and mention one risk.

Show the solution
  1. Contract review involves text, so use natural language processing. It can extract clauses, flag unusual terms and summarise obligations.
  2. Scanned KYC documents are images, so use computer vision. Optical character recognition reads the text and face matching can compare a photo with the document.
  3. Benefit: both tasks become faster and need less manual effort, with more consistent checks.
  4. Risk: errors or bias can wrongly flag or miss items, and personal data is processed.
  5. Safeguard: keep human review of flagged and rejected cases and protect the personal data.

Answer: Use NLP for contract review and computer vision for KYC document verification. The key risk is error and misuse of personal data, so human review and data protection safeguards are needed.

Example 2

The compliance team of a company uses a generative AI tool to draft a first version of a board resolution. Explain generative AI and advise on the risks involved.

Show the solution
  1. Define: generative AI creates new content, here text, from patterns learned from large data sets and a prompt given by the user.
  2. Benefit: it saves drafting time and gives a structure to start from.
  3. Risk 1: hallucination, so the draft may contain wrong references or facts.
  4. Risk 2: confidentiality, since company information typed into an external tool may be exposed.
  5. Risk 3: accountability, because the company and its officers remain responsible for the final document.
  6. Advice: use the tool only for first drafts, do not enter confidential data in public tools, have the company secretary verify every point against the law and records, and keep a usage policy.

Answer: Generative AI creates new content from learned patterns. It can speed up drafting, but because of hallucination, confidentiality and accountability risks, every draft must be verified by a qualified person and used under a policy.

Exam tips

  • Match technology to input first. Most case questions can be solved from that one step.
  • Always give both a benefit and a risk. Answers with only one side lose marks.
  • Use correct terms: hallucination, OCR, inference engine, knowledge base, RPA.
  • For governance questions, finish with human oversight and board accountability.
  • Keep each technology to a definition, one use and one risk if time is short.

Practice questions from Artificial Intelligence - Introduction and Basics

Key AI Technologies and Applications in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Key AI Technologies and Applications: frequently asked questions

What is generative AI in simple words?

Generative AI is AI that creates new content such as text, images or code. It learns patterns from large data sets and responds to a prompt. Its output can be wrong, so it must be checked.

What is the difference between NLP and computer vision?

NLP works with human language, written or spoken. Computer vision works with images and video. Both can be used together, for example when OCR reads a scanned document and NLP then analyses the text.

How is AI used in banking and finance?

Common uses are credit scoring, fraud detection, KYC verification, chatbots, algorithmic trading and robo-advice. Each use depends on a technology such as machine learning, NLP or computer vision. Banks still need human oversight and data protection.

How is AI used in corporate governance and compliance?

It can track regulatory changes, review contracts, screen transactions, and help draft documents. These tools support decisions but do not replace the responsibility of directors and professionals.