Why We're Watching AI Carefully Before We Commit

Business News Wales

27-06-2025

We haven't yet adopted AI in any widespread way across our HR practices at the University of the Arts London. But like many others in the sector, I'm watching its development with interest – and with a fair amount of hope for what it could help us achieve. While we're not using it now, we are thinking carefully about how it might fit into our future resourcing strategy.
One thing I'm clear on is that individuality still matters. When candidates apply for roles with us, we encourage them to write in their own tone of voice. We want to see who they are, not just how well they can mimic a job description. That's part of why we've stuck with human decision-making at every stage so far, even when managing several hundred applications for a single vacancy. That approach isn't without its challenges, but we've tried to manage it practically – for example, encouraging hiring managers to shortlist gradually rather than waiting until a role closes.
That said, I recognise the potential of AI to improve some of the more routine elements of recruitment. Our applicant tracking system provider is developing a new tool aligned to skills-based hiring, which we already use through our competency framework. If it works as well as the early version I've seen, it could help us manage volume and complexity without losing the human element. I don't see AI making final decisions for us but I can see it offering a helpful starting point. It might rank applications, surface potential matches, or highlight gaps to explore. The final judgment should still rest with people, but I'm open to tools that make that process more focused.
Beyond recruitment, I'm interested in AI's potential to support decision-making more broadly. Sometimes in HR, particularly when dealing with complaints or complex situations, it's useful to have something to help sense-check your thinking. Not to take over the process, but to confirm that the route you're taking is sound, or to suggest questions you haven't yet considered. I've already used AI tools in that way – not with confidential data, but to help frame responses or understand an issue more clearly when time is short and the pressure is high.
Like many professionals, I juggle a lot of priorities. There are always several projects, competing deadlines, and daily operational tasks. AI can play a role in helping me manage that. For example, it's already helped me to create structured plans and summaries more quickly, freeing up headspace for more strategic work. And that's where I think its real value lies – not in replacing HR roles, but in supporting us to focus more on the parts of our jobs that add long-term value.
Workforce planning is another area where I see potential. With the right data, AI could help identify where future risks lie, whether that's retention issues, capacity challenges, or the impact of organisational change. I've seen examples in previous roles where predictive analysis gave useful insight. When an organisation moved offices, we were able to forecast how many people might leave as a result of a longer commute. That kind of visibility is valuable. If tools can help us plan redeployment, identify skills gaps, or prepare for change in a more proactive way, that's a step forward.
There's also a possible benefit for individuals themselves. If you had access to a tool that monitored patterns in your organisation's performance and market conditions, you might be able to assess your own job security more realistically. Having gone through several rounds of redundancy in my career, I can see how helpful that kind of insight might be, offering early warning signs and helping people to plan ahead.
That said, I do have concerns. I don't want AI to dilute the human element of employee experience. We talk a lot about people-centric HR, and that has to remain the focus. There's a risk that, in automating too much, we lose some of the empathy and individual attention that people need, especially during periods of stress or poor mental health. I've experienced those challenges myself, and while AI might play a useful role in guiding someone through a difficult time, it can't replace genuine human support. The same goes for wellbeing tools more broadly. If done well, they might become part of a wider package, but they can't be the whole answer.
Looking ahead, I think we'll see a mix of outcomes. Some roles may change. Some tasks may be automated. But HR is a resilient profession, and I believe we'll adapt. AI won't replace us, but those who know how to use it effectively will likely be better placed to succeed. For me, the question isn't whether AI is coming into HR – it's how we make sure it's used well. If we get that right, it can be a powerful tool that empowers our work, rather than replacing it.
Listen to Mark discuss this and more in episode one of the DeeplearnHS podcast series, AI in Hiring & Workforce Strategy, here