Preparing for a generation of AI-literate employees  

This summer’s graduates are different. They enter the workforce as the first generation shaped by generative AI. They didn’t just study with these tools; they used these tools to solve problems, research and write.

 

At the same time, expectations from employers have shifted. Organisations are rapidly reshaping entry-level roles around AI, making critical thinking to evaluate outputs and understand AI’s opportunities and limitations more essential than ever before.

 

The real challenge is not whether graduates are work-ready, but whether employers are ready to support them.

 

 

A growing gap in expectations

Employer expectations around AI are rising sharply, even as graduates step into a challenging job market. Businesses are looking for candidates who use AI thoughtfully, responsibly and transparently.

 

However, many organisations are still figuring this out. Research shows that 61% of employers have no employees dedicated to working with AI, and only 11% offered any form of AI training in the past year. This creates a clear mismatch. Businesses expect new hires to demonstrate confident use of AI, yet are still defining what “good” looks like internally.

 

The nature of entry-level work is shifting, with the Institute of Student Employers noting that 87% of organisations expect AI to reshape early-career roles. However, a disconnect has emerged between businesses and higher education. Without defined standards for what employers expect from incoming talent, universities struggle to design relevant coursework. This leaves students uncertain about when and how to use AI, how to disclose its use, or where the boundaries lie.

 

In the absence of clear institutional guidance, students are largely teaching themselves. Seven in ten alumni say they developed their AI skills through experimentation. Only a small proportion have completed formal university courses focused on AI skills. Experimentation builds familiarity, but it doesn’t develop deeper capability. This creates an ecosystem where "AI slop" can make its way into professional workflows because no one is visibly learning together. Knowing how to generate a prompt is a world away from knowing how to challenge, validate, or responsibly apply it.

 

 

Rethinking onboarding for an AI-enabled workforce

Bridging this gap requires organisations to rethink how they bring early-career talent into the business. AI is not just another tool. It is reshaping the nature of work, placing greater emphasis on judgment, creativity and decision-making.  

 

Employers must move beyond the assumption that younger hires already “get” AI. Familiarity with tools does not translate into an understanding of their limitations or appropriate use in a professional context.

 

Effective onboarding needs to be comprehensive: 

• Role-specific examples: Show how AI should (and should not) be used in day-to-day work.
• Opportunities to experiment safely: Employees need to test approaches and build confidence without risk.
• Clear guidelines on responsible AI use: Employers need to define when tools can be used, where restrictions apply, and what accountability looks like.
• Training on risk and limitations: Employees need training on data privacy, hallucinations, bias, and reliability. 

Most issues arise from misunderstanding rather than deliberate misuse. Graduates may be comfortable producing AI-generated content, but they are often less aware of issues like bias or the implications of sharing sensitive information with third-party tools. Onboarding should focus on the skills that matter most. New hires need to learn how to identify unreliable information, handle data safely and explain their decisions clearly.

 

 

Shared responsibility

Preparing for an AI-enabled workforce is a shared responsibility. Universities will need to continue to embed AI into teaching in ways that build both practical skills and critical thinking. At the same time, employers should recognise that graduates are arriving with potential rather than fully-formed expertise.

 

AI awareness is just the baseline. Real value depends on whether organisations provide the structure, clarity and support needed to turn that familiarity into effective workplace performance. 

 

---

 

Annie Chechitelli is Chief Product Officer of Turnitin  

 

Main image courtesy of iStockPhoto.com and Thai Liang Lim