Graduates need AI skills – and the judgement to use them

Graduates are entering a labour market in which artificial intelligence is changing how work is produced, priced and delivered.

AI and big data are among the fastest-growing workplace skills, but employers continue to value human capabilities such as analytical thinking, creativity, resilience and collaboration.

Leon Smalberger, Chief Executive Officer of the Academic Institute of Excellence (AIE), says graduates need to understand how to apply AI effectively and responsibly rather than simply knowing how to operate a tool.

They must combine AI capability with professional knowledge, creativity, judgement and problem-solving.

AI can lower barriers to entrepreneurship

AI is making some services more affordable to produce, creating opportunities for graduates, entrepreneurs and small teams that do not have the resources of larger organisations.

“Employers are increasingly seeking people who can put AI to work effectively and responsibly within their roles,” Smalberger says.

“Those same capabilities are also lowering barriers to entrepreneurship for graduates, allowing small teams to compete in ways that were previously only possible for much larger companies with greater headcount or budgets.”

A graduate could, for example, use AI to reduce the time required to develop visual training material for a small manufacturer.

The technology might assist with animation, images or production. However, the designer would still need to understand the equipment, interpret the brief and decide how to explain a complicated process clearly and accurately.

A developer could similarly use AI-assisted coding to create an equipment-tracking and customer-approval system for a repair workshop that still relies on paperwork and WhatsApp.

The developer would remain responsible for testing the system, protecting customer information and ensuring that workshop employees could use it safely and effectively.

Small teams can deliver more

Smalberger says AI can help graduates establish one-person businesses by supporting tasks that might previously have required several employees or outside specialists.

“We’re already seeing fresh graduates launch one-person businesses and use AI to handle essential tasks that would previously have required several employees,” he says.

“They can build a functioning business with minimal resources, establish a steady income and bring in their first employees when the revenue can sustain them.”

This does not mean that AI removes the need for expertise. Graduates still need to understand the problem they are trying to solve, assess whether the proposed solution is appropriate and take responsibility for the final result.

The opportunity lies in using AI to extend professional capability rather than allowing it to replace professional judgement.

AI mistakes can become costly

AI can also create financial, operational and reputational risks when its outputs are accepted without scrutiny.

“A low- or no-cost system can be the most expensive and reputation-damaging decision you make early in your career or entrepreneurial journey,” Smalberger says.

“A single serious AI misinterpretation or hallucination could derail a product launch, trigger costly accounting mistakes or spark customer backlash.”

An AI hallucination occurs when a system produces false, misleading or fabricated information and presents it as though it were accurate.

Students therefore need to learn how to verify outputs, identify unsupported claims, test AI-assisted work and recognise when a task requires additional expertise or human approval.

Professional knowledge remains the safeguard

Subject knowledge helps graduates detect errors, understand why an AI-generated response is inadequate and provide more precise instructions.

Students should learn to analyse mistakes in prompting, data and system design, explain why they occurred and refine their work under appropriate supervision.

They also need to judge how much human oversight a task requires. Drafting ideas for a presentation, for example, does not carry the same level of risk as producing financial calculations, legal guidance, medical information or software controlling important operations.

AI education should consequently include ethics, privacy, data quality, bias, accountability and responsible decision-making alongside technical proficiency.

What an AI-ready qualification should provide

Integrating AI into a qualification involves more than adding a short module on prompting.

Students need opportunities to use AI within the context of their chosen professions, compare its output with reliable evidence and apply professional standards to the final work.

AIE’s Higher Certificate in Information Technology in Artificial Intelligence and Bachelor of Information Technology in Data Science, for example, combine computing principles with practical analysis, machine learning and data-science ethics.

Students considering a qualification should examine whether its AI content includes:

• practical work linked to real professional problems;
• strong foundations in the main discipline;
• verification and testing of AI-generated outputs;
• ethics, privacy, bias and responsible data use;
• opportunities to develop creativity and critical thinking; and
• assessment of the student’s own reasoning, not only the completed product.

“Very few careers or industries will escape the influence of AI,” Smalberger says.

Students who learn to combine AI capability with sound professional knowledge and human judgement will be better prepared to adapt as jobs change, identify new opportunities and build careers or businesses of their own.