Job applications are the visible tip of the AI iceberg for unions

Have you tried hiring anyone for a job recently? If you have, you’ve likely been frustrated by having to wade through dozens of job applications generated by ChatGPT, or other generative AI tools

This frustration is increasingly common across the union movement. Union HR officers at a recent Unions21 roundtable were already finding it hard to handle the volume of AI-generated applications. As AI tools become more accessible, job applicants are using them to create their applications - sometimes without even doing a basic review.

But rather than just being an irritation for recruiters and HR teams, these AI-generated applications point to a much bigger challenge that unions need to address. That’s because they are just the most visible sign of how AI is already changing how people engage with organisations and each other.

Why people use AI for job applications

But consider this: the number of applicants using AI to write their job applications is likely to grow, rapidly. At the moment, Hays research found only 18% of candidates say they have used an AI tool when completing a job application (rising to 24% of 18-24-year-olds). When asked if they’d consider use AI in future, only 36% reject the idea outright.

What’s the attraction? There are two important reasons why this is happening…

The first reason is that candidates they think it helps them get jobs. 49% of those who have used AI in their job applications feel they have been more successful.

The second is more interesting. AI makes it easier to overcome the pain of application processes. Our recruitment processes rely on friction to avoid people sending in well-developed but somewhat speculative applications. Applying for a job used to have real costs and risks, but the cost of applying with AI is now near zero time or effort. Have a watch of this video on Instagram to see what I mean. It shows how very easy creating powerful job applications can be, even for the hardest companies.

Unions can't just say no to AI use

Our first instinct as recruiters might be to try to block AI use - indeed, some organisations are already adding "no AI-generated applications" clauses to their recruitment processes. But this approach faces several problems:

1. It's difficult to detect AI writing reliably - the AI detectors on the market are notorious for false positives, and can even introduce bias.

2. The bans could create barriers for candidates who use AI as an accessibility tool

3. Banning fails to engage with the reality that AI skills may soon be essential in many roles.

As one participant in the roundtable noted: "If we ban AI applications outright, we might be creating barriers for people who need these tools due to disabilities. We need to think this through carefully."

A different approach: engaging with AI

The challenges posed by AI-generated job applications might seem like a headache for recruiters.

But they're actually an opportunity to start developing the practical expertise unions will need in an AI-enabled world. Instead of trying to block AI use, unions have an opportunity to model responsible engagement. This could mean:

1. Being explicit about allowing AI use in applications while requiring transparency of candidates about how they used them.

2. Developing clear guidelines that reflect our values about appropriate AI use that makes processes fairer for all.

3. Investing in building our own AI capabilities, so we can are better placed to evaluate a candidate’s use of AI.

4. Redesigning recruitment processes to emphasise human interaction and work with, rather than against, technological change.

5. Sharing learning across the movement about what works.

Ultimately, how unions handle AI in recruitment sends important signals about our broader approach to technological change. If we want to effectively represent workers in discussions about workplace AI use, we need to develop practical experience in managing these tools ourselves.

This means developing expertise not just in the technology, but in the human and organisational challenges it creates. As another participant in our HR roundtable put it: "We need to think about the culture we're creating - if people feel they need to hide their AI use from the start, what does that mean for transparency later?"

This post draws on discussions from the Unions21 HR roundtable series and our ongoing research into AI's impact on union operations.

Image: Jamillah Knowles / Better Images of AI / Data People / CC-BY 4.0