When you ask ChatGPT to draft a job description for a Business Director, you might be impressed at first glance — it provides a comprehensive description with the basics, such as key responsibilities and qualifications.
“This role requires a strategic thinker with a proven track record of driving business growth, optimising operational performance and fostering a culture of excellence. The ideal candidate will possess strong leadership skills, business acumen, and the ability to navigate complex business environments and negotiate contracts.”
But take a closer look. ChatGPT has fallen at the first diversity, equity and inclusion (DE&I) hurdle. Traditionally masculine coded words are used throughout – “proven”, “drive”, and “strong”.
How about a different GenAI tool, with a different output format? We asked Copilot to create an image of a Business Director. Here, the gender bias is stark.
GenAI is not data processing in a vacuum. It holds our existing human bias, pulled directly from the content it’s trained on. This creates fundamental questions about the legitimacy of outputs.
But there’s hope! We’re now seeing a rise in AI tools designed specifically to combat bias. Could AI be the newest tool HR leaders can use to overcome fundamental challenges to improve workplace diversity, equity and inclusion (DE&I)?
We explore whether AI is the latest foe, or ally, to workplace DE&I efforts.
Tackling unconscious bias
HR leaders are using AI tools that can analyse text, highlight non-inclusive language, and suggest alternatives to remove gendered wording from their applications. Though simple, this can have stunning results. Danske Bank increased the number of qualified female applicants by 81% using Develop Diverse (Springwise), a writing platform that detects biased words and sentences and suggests inclusive and non-stereotypic alternatives.
What about further down the hiring journey?1 This is not a solution-all answer—recruitment teams still need to use human eyes to ensure that patterns of bias are not emerging.
Understanding and personalising through AI
Organisations know the benefits of building an inclusive workplace, but the journey to get there is not as simple. Often, inadequate data and analytics make it difficult to interpret the root causes underpinning employee engagement and retention trends. Using AI analytics tools, organisations can go beyond spotting bias patterns and get to the source of the issue.
The ability to scan large data sets using AI and create predictions opens up possibilities for more personalised (and therefore, inclusive and effective) learning pathways. The University of Sydney has used Smart Sparrow to create adaptive learning pathways, improving academic performance through a learning experience that adapts to students’ individual needs in real-time.2
Instead of perpetuating existing bias, let’s combat bias – at pace
As the examples above show, AI can help us to address unconscious bias, but the paradox is that it can also perpetuate existing bias. Bad data in = bad data out. We have seen organisations grapple with this issue – Amazon had to scrap their AI-driven recruitment tool after realising that it reinforced gender bias rather than eliminating it.
AI is a powerful enabler for DE&I, but it is not a universal remedy. Working with AI can help us overcome our own unconscious bias, but we must remember that AI itself does not have a moral compass. That’s our job, which is why we believe that as humans, we can be better with AI and should always be better than AI.
Four steps to start exploring AI opportunities for your organisation
1. Set up your AI labs
A cross-functional team, including HR and tech practitioners, is a great way to collaboratively explore, test and pilot where AI can make an impact.
2. Look back at your DE&I values and mission, what are you trying to achieve?
Implement AI with purpose. Before reviewing your existing processes (e.g. recruitment, L&D) and systems to understand how AI can be integrated, come back to your ultimate why and let that lead you.
3. Get experimental
Even if our ambitions are bold, start small and grow from here—harnessing the agility of your cross-functional team, running small AI experiments to test new functionality, and monitoring the results closely. Does this technology bring you closer to your mission?
4. Don’t forget the human element
Establish governance and checks to ensure your efforts continue to improve DE&I, and you – and the AI tools – learn and improve as you go.
WORTH THE EFFORT
Building data management maturity is a continuous, incremental, and sometimes arduous process, but high-quality, secure data is fundamental to maintaining the trust of your customers and enabling leading-edge analytics to inform decisions.
Sign up to The People Strategist, our regular Gate One newsletter that helps you enable your people to be the best they can be.
