Artificial Intelligence – The Future of HR?

Artificial intelligence (AI) is the hot technology in HR. These days, it’s almost impossible to find an HR technology vendor who doesn’t claim their product is ‘powered by AI’. It’s already proliferating in our daily lives: I’m amazed, when I ask Alexa “What it’s like in London today?”, that she knows I’m asking about the weather. But when it comes to applying AI to the messy business of managing the relationship between human beings and their employer, what’s the role of AI? Should we in HR see it as an existential threat or an opportunity to get the machines to take care of routine, repetitive work, while we focus on adding value to the business? To what extent will AI shape how we view the purpose and role of HR?

First, let’s be clear about what we mean by AI. It’s technology that can do the types of things that traditionally have required human intelligence, such as learning, problem solving, and pattern recognition. For example, Google DeepMind has developed a programme, AlphaGo, that taught itself the complex Chinese strategy game Go, and beat the world number one player.

AI is being applied in HR today, not only to handle routine tasks such as scheduling interviews, but also increasingly to do things that until recently were only possible with human judgement, such as screening job candidates. Although sometimes vendor claims are way ahead of what these tools can do in reality, AI is becoming an intelligent decision-support tool. Here are some of the applications we are seeing:

  • Personalising the employee experience. Netflix and Amazon have become highly sophisticated at building algorithms that make recommendations about content or products we might be interested in as consumers, based on previous choices. A challenge for HR is to deliver a similarly personalised experience at work. Learning tools such as Degreed serve up content based on an individual’s browsing history, and what others with similar profiles have viewed.
  • Recruitment. CV scraping gives recruiters access to a global talent pool. Text analysis engines can infer an individual’s personality profile or preferences by analysing only a few hundred words in an email. It’s no longer necessary for a human to read every one of the hundreds of thousands of CVs some large companies receive annually. Applicant sifting tools such as IBM’s Watson Recruitment engine can read CVs, interpret and infer candidates’ skills, and compare an individual’s profile with a competency library to predict which role would be the best match. HireVue, the video interviewing platform, can read speech patterns, voice-generated emotions and facial gestures, and detect behavioural attributes that predict on-the-job performance. Large companies have reduced the cost and time to hire by getting machines to shortlist candidates, only getting humans involved in assessing candidates at later stages.
  • Talent management and workforce planning. Cisco is developing a Talent Cloud that helps the company do more agile talent planning and helps individuals better plan their careers. It compares individuals’ skills against the requirements of their current and potential future roles, suggests training courses or project assignments, and maps potential future career paths. The aggregate data helps the company work out where skills gaps might need to be plugged by recruitment, training, or acquiring another company.
  • HR service provision. Bots are increasingly replacing HR advisors. They are being deployed to ask candidates whether they have the right to work in a particular country, schedule training courses or answer questions on the company’s maternity policy, for example.

We see AI offering a number of potential benefits for the HR function.

  • For over twenty years, CRF has been urging HR to become more evidence-driven. AI has the potential to help HR rely less on gut feel and be more systematic in decision-making. Machines don’t get tired or make poor decisions before lunch.
  • By systematising much of the lower-value work that HR has historically been responsible for, this should free up capacity for HR to support the business in delivering what’s needed from a people and organisational perspective to execute the business strategy – provided we develop the business skills and credibility to take on this work.
  • AI can also help organisations become more diverse. Machines are programmed to select the best candidates, regardless of gender, skin colour, or background. However, as algorithms are typically designed using historical data, we do need to take care to test and refine them to eliminate in-built bias.

However, before we get too excited about the possibilities, we need to recognise that we are probably somewhere near the top of the hype cycle for AI. We need to both be realistic about the limitations of the technology and recognise the background work that’s necessary to uncover the benefits.

  • HR will need to develop new skills in a technology-driven world. While business and data-savvy HR professionals will thrive, certain breeds of HR people – those who make decisions based on gut feel or are terrified of numbers – will become an endangered species. How are we supporting our own HR function to acquire the necessary skills? Are we investing in good change management for ourselves?
  • We have to be an educated buyer. Technology invariably turns out to be less effective than expected. We have to get good at due diligence, asking the right questions, and getting under the skin of the technology to understand what it can – and more importantly what it can’t – do.
  • Are we thinking about how our people policies need to evolve in a world where the employee experience is highly personalised? Is it still relevant to have one-size fits all policies for compensation, performance management, career development etc?
  • We need to invest in improving the quality of people and performance data, and get better at collaborating with other functions such as marketing and technology to build solutions that address business rather than HR needs.
  • We have to develop appropriate ethical frameworks and manage risk. Technology is advancing at pace, but the development of ethical norms is not keeping up. We also have to be careful that the algorithms do not programme in unconscious bias.

In summary, AI is something we should be embracing in HR. It may not put HR out of a job, but it will challenge us to think differently about our role and how we support our businesses. It’s an opportunity for us to become more business-relevant.

Back to top