How get the most out of AI in HR

Technology is evolving at a rapid pace, and businesses must ensure their workforce has the skills needed to keep up. The swift rise of artificial intelligence (AI), and generative AI in particular, is transforming the way many organisations operate. According to a recent report, 82 per cent of enterprise data and IT leaders say that AI is already impacting what their organisations can achieve. And many companies anticipate significant changes to come.

Yet a historical reliance on hiring only “specialist” talent, such as those with specific coding and programming skills, can hold many organisations back from capitalising on the valuable data-led insights that AI will provide. Indeed, the application of AI is already expanding beyond tech and IT teams; marketing and communications, for example, were highlighted in the report as business areas expressing high demand for AI.

Despite the well-documented AI skills gap, where demand has far outstripped supply, many businesses still prioritise recruiting for roles with highly sought-after technical skills. However, robust oversight mechanisms which allow a “human-in-the-loop” or “human-in-command” approach to AI innovation are vital for ensuring that a human can hold the steering wheel on AI to audit and guide the AI's operations while making use of its full capabilities. Furthermore, soft skills will be required for collaborating with and questioning the outputs of intelligent systems and this requires humans with use case knowledge and governance of the AI results.

Therefore, this disconnect in skill priorities means organisations should rethink their hiring strategy regarding the skills required to deliver data insights and decision intelligence through AI. Rather than recruiting specialists, businesses should consider hiring talent with AI-compatible skills while upskilling mid- and later-career workers with the business skills and empowering them to work with AI.

Balancing hard and soft skills

Upskilling doesn’t mean all employees should become data scientists. After all, the art of extracting insights from enterprise AI applications doesn’t depend entirely on technical skills. Rather, it’s essential to build a multi-skilled workforce that feels confident in the use of AI to address different use cases, combining extensive domain knowledge with the ability to “speak data”. It’s encouraging, then, that almost three-quarters of leaders (72%) believe it is more important for employees to be multi-skilled than specialised in a specific area.

Yet, despite growing recognition of the importance of creativity and critical thinking in collaborating with and questioning the outputs of AI applications, these were among the lowest-ranked skills in terms of priority, according to the report. In fact, “hard” skills, including expertise in emerging technologies like AI, data analysis, and data mining, all ranked as more in demand than “soft” skills, such as digital literacy, team leadership, and time management. With AI, its crucial the business is asking the right questions ahead of implementing generative AI.

The ability to scrutinise any answers given by AI is essential to ethics and governance. So, when you consider that three-quarters (73%) of decision-makers admitted concerns around AI-produced answers in their organisation, the low priority given to critical thinking in the talent assessment process is something of a concern. And when 61 per cent of decision-makers believe that creativity is the top skill humans will supply in a world shaped by AI, there’s clearly a vital need for employees to retain “human” qualities at the same time as acquiring new AI-based skills. It’s important to remember that humans will have views into the impact on the business that are not obvious to AI, so humans will remain key to asking the right questions and defining the scope of AI apps.

Data-literate culture

In practice, upskilling should equip employees with the knowledge they need to ask the right questions to input into AI applications like Large Language Models (LLMs) to receive helpful answers. These questions can enable LLMs to deliver far more effective outputs when they’re informed by domain-specific expertise; this is why upskilling mid- to later-career workers who hold this specific expertise is ideal.

As we’ve seen, all of this – the development of AI skills, combined with domain-specific business context – should be paired with an emphasis on transferable soft skills to create an internal data-literate culture. By assessing an employee’s current technical and soft skills, HR can then create a bespoke training programme that compliments wide-reaching skills training with accessible tools that democratise employee access to data.

Such tools are widely available today. The ascendancy of no- and low-code self-service platforms means businesses can expand the reach of AI and machine learning applications to where they’re needed across the organisation. The growing use of these platforms in many learning institutions across a range of subjects demonstrates perfectly how AI is no longer the sole domain of data scientists – it must be used by (almost) anyone for (almost) any use case.

Flexible and dynamic

It shouldn’t be news to anybody that AI is evolving rapidly, and businesses looking to capitalise on this need to get their technology rollout right. A flexible and dynamic approach to upskilling existing employees and hiring talent with AI-related skills forms the basis of a winning strategy.

Scaling any organization will require the provisioning of ongoing training for both new and experienced AI application users to understand how to manage new data types like visual and audio inputs, as well as applications like computer vision and natural language processing (NLP). At the same time, of course, the retention of soft skills such as critical thinking will be vital in managing the ethical considerations, data privacy and governance of an organisation’s AI applications.

Only by embedding data literacy and analytics upskilling in employee growth strategies can business leaders create a data-literate workplace in which everyone is able to use AI to deliver data-driven insights at the speed of business.

‍