AI Is Rewriting Graduate Work. Are We Preparing Students?
Artificial intelligence is "without doubt" transforming parts of the global labour force. That was the clear message from Prof Duncan Ivison, president and vice chancellor of The University of Manchester, during his visit to the UAE.
He noted that while AI will create new roles, tasks once handled by graduate hires in consultancy, legal, and analysis are being taken on by models - often built in-house. "I don't think we can say AI is responsible for everything," he added, but the direction is obvious: roles are being eliminated or changed, and new tasks are emerging.
His core question is our challenge as educators: "How do we make sure people are prepared for those changes?"
Why this matters for education leaders
Entry-level work is shifting fast. Automation, tax and investment pressures, and organizational adoption of AI are reshaping how teams operate. Graduates now need AI fluency, ethics, and the ability to collaborate with tools - not compete with them.
Students are asking for it. They want useful AI skills and to learn how to use AI responsibly, with a clear sense of values. That should be the baseline across programs, not an optional extra.
Practical moves to make this semester
- Build AI literacy in every program: Teach prompting, data basics, verification, and model limitations. Use discipline-specific labs (e.g., legal drafting reviews, financial analysis with model audits, qualitative coding support for social sciences).
- Teach Responsible AI: Anchor to societal values: fairness, accountability, transparency, privacy, and safety. Use case studies on bias, consent, and misinformation. Make "appropriate use" policies visible and simple.
- Rethink assessment: Combine AI-permitted tasks with oral defenses, process logs, and source portfolios. Grade reasoning and critique, not just outputs. Be explicit about what AI use is allowed.
- Upskill faculty and staff: Fund short sprints for experimentation and share templates across departments. Build a peer community that swaps prompts, rubrics, and workflows.
- Protect equity: Provide campus access to licensed tools, GPU time, and support for students with disabilities. Offer low-cost or open-source options where possible so access isn't paywalled.
- Align careers and industry projects: Map which tasks in each field are changing and set up employer-backed projects that reflect them. Prepare students for roles that merge domain expertise with AI fluency.
- Modernize research workflows: Treat generative tools as accelerators - "the equivalent of having a postdoc or a couple of PhD students," as Prof Ivison put it - while training for validation, reproducibility, and data ethics.
- Set clear governance: Create a cross-functional AI group covering procurement, privacy, IP, academic integrity, and risk. Keep guidance lightweight and updated.
For context, the IMF estimates AI will affect roughly 40% of jobs globally - replacing some tasks and complementing others. That's exactly why practical readiness beats abstract debate.
Global context: mobility, hubs, and what's changing
Student migration is reshaping higher education. There's a dip in international students heading to the US and caps in other regions. Yet demand for strong programs remains steady - and hubs like Dubai continue to grow.
The University of Manchester opened its Dubai centre in 2006, offering MBAs and programs in educational leadership and finance. More than 2,600 students have graduated, and many stay to build careers in Dubai and Abu Dhabi. The 2025 class includes 135 students from around the world.
We're also seeing more learners use the MBA as a springboard into startups or entirely new roles. That's a cue to widen employer partnerships and flexible pathways.
Strategy and funding: building for 2035
Founded in 1824 during the Industrial Revolution, The University of Manchester has about 44,000 students, 26 Nobel laureate alumni, and ranks sixth in the UK and 46th globally in the 2025 academic ranking. Its new £400 million global fundraising and volunteering campaign aims to push forward education and solutions to major challenges.
The "Manchester 2035" strategy leans on global centres like Dubai to address grand challenges across technology, climate, migration, and entrenched inequality - and how these issues intersect. The approach is unapologetically interdisciplinary: material scientists with economists, philosophers with health researchers, and computer scientists with clinicians.
A 90-day action plan for your institution
- Set policy clarity: Publish simple AI-use guidelines for students and staff. Include examples and a one-page "what's allowed" chart.
- Run pilot modules: Add one AI-supported assignment in at least three high-enrolment courses (business, computing, social sciences). Measure outcomes and iterate.
- Faculty sprint: Host a two-week micro-lab where instructors adapt one assessment with AI, plus a show-and-tell at the end.
- Equity check: Ensure campus access to tools, provide training drop-ins, and set up a "no personal credit card needed" option.
- Employer-backed projects: Launch two capstones where students use AI on real datasets or briefs from local partners.
- Skill pathways: Offer short credentials for AI literacy by job family (teaching, admin, research). If you need a curated catalog to get moving, explore AI courses by job.
The bottom line
AI is changing work faster than our old course maps. As Prof Ivison put it, AI is transforming the labour force, and "we don't want to leave people behind."
Prepare people for the work they'll actually do: teach responsible use, refit assessments, upskill faculty, and protect access. That's how we keep opportunity open while the task mix shifts under our feet.
Your membership also unlocks:
