Renuka Kamath

Bridging the AI talent gap between university curricula and industry demand | Renuka Kamath | Professor of Marketing, Associate Dean of Full-time Programmes | SPJIMR

Renuka KamathThis is the paradox confronting academic institutions and industry. The AI talent gap is often described as a shortage of people who understand AI. I believe the problem is more nuanced. While the younger generation is adopting AI tools at an extraordinary pace, the real gap lies between using AI and working intelligently with it. Industry increasingly needs the latter.

Consider a young marketing graduate. She can get AI to create a customer persona, analyse reviews, and draft a campaign in minutes. This is impressive, but will she be able to respond to questions like Why do you trust this analysis? And what assumptions did AI make? What would you refuse to delegate to it? She needs to develop the ability to recognize when the persona she created through AI is based on flawed assumptions. She will need to be prudent about which customer data to share with the AI platform. Would she be able to defend the recommendations before a sceptical client? And most importantly, can she recognize when not to use AI? These are not merely technology questions but questions of using judgement correctly.

Don’t add AI to the curriculum. Put AI through it

Academic institutions have traditionally organised knowledge into disciplines such as Finance, Marketing, Operations, and HR. AI does not respect boundaries. It is embedded in each function. Hence, it cannot be treated as a subject to be added to an already overcrowded curriculum. Doing so misses the point. It needs to be a layer across the curriculum. A Finance student should cross-examine an AI-generated valuation, a Marketing student could use AI to analyse customers, an HR student could use it to examine algorithmic hiring recommendations for bias, and an Operations student could work with AI-generated forecasts and question them.

All these points lead to another important shift. From students primarily producing answers to learning to evaluate answers. For years, education rewarded the student who produced the right answer or the correct solution. This is becoming inexpensive and easy in the workplace. The premium will increasingly be on asking the right questions, interrogating evidence, identifying hallucinations, understanding context, and, finally, exercising judgement about whether the output is commercially viable and ethical. Still, it isn’t that simple, as the employer has a role too!

Industry plays a role too

‘Industry unready’ graduates being produced by academic institutions is the usual refrain of the corporate sector. The truth is, industry demand is changing faster than a university can redesign and deliver courses. Employers must therefore become co-creators rather than only customers of education by co-teaching, providing anonymised live projects, providing real data sets, providing faculty immersion in their organisations, etc. Practitioners’ constant engagement and help to educational institutions will help refresh curricula.

We may be measuring AI readiness incorrectly

This temptation to measure preparedness through the number of students trained in AI, the courses launched, or certificates completed must change. A better test might be surprisingly simple. Give a student a messy business problem, access to AI tools, and incomplete information. Then observe: What questions does she ask? What does she delegate to AI? What does she retain for herself? Does he challenge the output? Can she combine machine intelligence with customer, organisational, and cultural context? And can she take responsibility for the final decision? This might be much closer to what tomorrow’s workplace will demand.

The AI talent gap can be closed by redesigning the relationship between knowledge, technology, and human judgement. Educational institutions will have to prepare students not merely to compete with AI nor simply to use it, but to know what to delegate to the tool, what to question, and what must remain fundamentally human. Moreover, particular AI tools are fashionable from time to time. Tools will change. The deeper capabilities of problem framing, domain knowledge integration, critical thinking, ethical reasoning, communication, and judgement will endure. These may ultimately be the most valuable skills of all. The question to ponder is how we move from producing graduates who are not just AI-skilled but AI-wise.

Leave a Reply

Your email address will not be published. Required fields are marked *