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

Benefits, Risks and Challenges of AI for CS Professional

Updated 11 October 2026 · Fact-checked

AI brings speed, accuracy, scale and cost savings, but it also carries risks: algorithmic bias, opaque black-box decisions, privacy loss, security attacks, unclear accountability and job displacement. To answer an exam question, list the benefits briefly, explain each risk with a business example, and close with practical controls such as governance, audits and human oversight.

Understand Benefits, Risks and Challenges of AI

Artificial Intelligence (AI) means systems that perform tasks that normally need human judgment, such as predicting, classifying, recommending or generating content. Companies use AI because it works on large data, at high speed, without tiring.

Benefits fall into a few groups. AI automates repetitive work. It improves accuracy in tasks like document review and fraud detection. It supports better decisions by finding patterns in data. It works round the clock, scales to millions of users, and can cut costs. It also enables new products, such as chatbots and personalised services.

Risks come mostly from how AI learns. Algorithmic bias arises when training data is skewed or incomplete, so the model gives unfair results to some groups. For example, a loan model trained on past approvals may disfavour applicants from certain areas. The black box problem means that complex models, such as deep neural networks, give outputs that even their developers cannot easily explain. This is also called lack of transparency or explainability.

Other risks follow. Privacy: AI needs large amounts of personal data, which can be over-collected, misused or re-identified. Security: models can be attacked through poisoned training data, manipulated inputs or theft of the model, and AI can also be used to create deepfakes and phishing. Accountability: when an AI decision causes harm, it is unclear whether the developer, the deployer or the user is responsible. Job displacement: routine roles may shrink, while demand grows for new skills.

The challenge for a company secretary is to help the board capture the benefits while managing these risks. This means governance policies, data quality checks, human oversight, impact assessments, audit trails and compliance with applicable law, including the Information Technology Act, 2000 and data protection requirements.

Key rules to remember

Core benefits of AI: Automation + Accuracy + Speed + Scale + Insight + Availability; Use as a checklist for the advantages part of an answer.
Core risks of AI: Bias + Opacity + Privacy + Security + Accountability + Job displacement; Cover all six in a full answer, with one example each.
Source of algorithmic bias: Biased or incomplete data (or design) → biased model output; Bias enters through data, labelling choices or model design, not only through the algorithm.
Black box problem: Complex model → output without an explainable reason; The remedy is explainability, documentation and human review.
Risk response: Identify → Assess → Control → Monitor; Use this to structure the mitigation part of any answer.

How to solve Benefits, Risks and Challenges of AI questions

Use this method for any descriptive question on AI benefits, risks or ethical issues.

1Read the question and note the verb: discuss, explain, list, advise or evaluate. Check whether it asks for benefits, risks, or both.
2Define AI in one line and, if the question gives a scenario, name the AI use in it (for example, credit scoring or recruitment).
3State the benefits in a short list, each with a one-line business reason.
4Take each risk in turn: define it, explain why it arises, and give a short example. Keep to the facts of the case.
5Link the risk to a legal or governance concern, such as privacy, data security under the IT Act, 2000, or board responsibility. Cite a section only if you are sure of it.
6Give controls: data quality checks, bias testing, explainability, human oversight, security safeguards, clear ownership and audit trails.
7Conclude with a balanced view: AI is useful when governed responsibly. For case questions, give a clear recommendation.

Quickest way: Benefit–Risk–Control (BRC) in three blocks

When to use it: Use when time is short or when the question is a 5-mark or 6-mark short note.

Write two or three benefits in one line each.
Write the risks as a list of six words: bias, opacity, privacy, security, accountability, jobs. Add one line of explanation for each.
Write three controls: governance policy, human oversight and regular audit. Add a one-line conclusion.

Common mistakes in Benefits, Risks and Challenges of AI

Writing only the risks and ignoring the benefits, or the reverse.
The topic title has three words, but students focus on the one they remember best.
Fix: Always give a short benefits block, then risks, then controls. Keep the balance in your conclusion.
Confusing algorithmic bias with a coding error.
Students assume bias means the programmer made a mistake.
Fix: Explain that bias usually comes from skewed training data, poor labelling or design choices, so a model can be technically correct and still unfair.
Defining the black box problem as a hidden security flaw.
The word 'black box' sounds like a hacking term.
Fix: Define it as the inability to explain how a model reached a result. Link it to transparency and explainability.
Giving generic points with no example.
Students memorise lists instead of applying them.
Fix: Attach a short Indian business example to each risk, such as a bank, an NBFC, a recruiter or an insurer.
Stating that AI is fully regulated by a specific named AI statute, or quoting section numbers from memory.
Students try to add legal detail to earn marks.
Fix: Refer to the law in plain words and cite a section only when you are certain. Say how existing laws such as the IT Act, 2000 apply to data and security.
Ending with no controls or recommendation.
Time runs out after the long risk section.
Fix: Reserve the last few lines for controls and a conclusion. Case questions expect advice.

Worked examples

Example 1

Aarav Finserve Ltd., an NBFC, uses an AI model to approve small loans. It finds that applicants from certain districts are rejected far more often, and the credit team cannot explain why individual loans were refused. Identify the risks and advise the board.

Show the solution

Provision: the facts show two AI risks. Rejections concentrated in certain districts point to algorithmic bias. Inability to explain individual refusals is the black box problem.
Analysis of bias: the model was probably trained on past lending data that under-represented or penalised those districts. The model repeats that pattern, which can lead to unfair treatment and reputational and regulatory risk.
Analysis of opacity: without explanations, the company cannot tell a customer why a loan was refused, cannot check fairness and cannot defend the decision to a regulator or court.
Related risks: the model uses personal data, so privacy and data security duties apply. Accountability is also unclear because no one owns the decision.
Advice: test the model for bias on location and other groups, review and balance the training data, adopt explainable models or explanation tools, and keep a human reviewer for rejections.
Advice on governance: assign a named owner for the model, keep logs and documentation, audit the model regularly and report to the board.

Answer: The facts show algorithmic bias and a black box problem, with privacy, security and accountability concerns. The board should order bias testing, improve data quality, require explainability, introduce human review of rejections, fix ownership and audit the model regularly.

Example 2

Discuss the benefits and the risks of using AI in a company's recruitment process.

Show the solution

Introduce: AI screens CVs, ranks candidates and schedules interviews.
Benefits: it saves time on large volumes of applications, applies the same criteria to everyone, reduces cost and can improve the match between candidate and role.
Risk of bias: if the model learns from past hires that favoured one profile, it may screen out equally capable candidates.
Risk of opacity: candidates and HR may not know why someone was rejected.
Risk of privacy and security: CVs hold personal data that must be collected for a clear purpose and protected from breach.
Accountability and jobs: the company stays responsible for hiring decisions, and HR roles may shift from screening to oversight.
Controls: regular bias audits, human review of final decisions, clear consent and data protection practices, and vendor due diligence.

Answer: AI in recruitment gives speed, consistency and cost savings, but it can bring bias, opacity, privacy and security risks and unclear accountability. The company should keep human oversight, audit the tool for bias, protect candidate data and remain responsible for final decisions.

Exam tips

Structure every answer as benefits, risks, controls, conclusion. Examiners reward a clear flow.
Use the six risk words (bias, opacity, privacy, security, accountability, jobs) as headings in your answer.
In case questions, spot the risk from the facts first, then name it. Do not list all risks without applying them.
Add practical points such as board policy, audit trail, human oversight and vendor checks, since the paper rewards compliance and drafting detail.
Mention a statute or section only if you are sure of it. A correct plain-words rule scores better than a wrong citation.

Practice questions from Artificial Intelligence - Introduction and Basics

Benefits, Risks and Challenges of AI in other exams

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

Benefits, Risks and Challenges of AI: frequently asked questions

What is algorithmic bias in AI?

Algorithmic bias is when an AI system gives systematically unfair results for some people or groups. It usually comes from skewed or incomplete training data, poor labelling or design choices. It can be reduced by testing, better data and human review.

What is the black box problem in AI?

It is the difficulty of explaining how a complex model, such as a deep neural network, reached a particular output. This lowers transparency and makes it hard to audit or challenge decisions. Explainability tools and documentation help.

What are the main advantages and disadvantages of AI for the CS Professional exam?

Advantages are automation, accuracy, speed, scale, better decisions and round-the-clock availability. Disadvantages are bias, opacity, privacy loss, security threats, unclear accountability and job displacement. Write both with examples and add controls.

Who is accountable when an AI system causes harm?

There is no single answer, as it depends on the facts, contracts and applicable law. Developers, the company that deploys the system and its users may all have roles. Companies should assign clear ownership and keep records of how the system was built and used.

How should a company manage AI risks?

Set a governance policy, test data and models for bias, keep humans in the loop for important decisions, secure the data and models, and audit regularly. Report significant AI risks to the board.