The Prevalence of AI in the Workplace
Let’s start by taking a look at the scale of Artificial Intelligence (AI) within the broader workplace context. AI has been ranked Number One on the 2020 Society for Industrial and Organizational Psychologies (SIOP) Trends in the Workplace, with a 2019 Harvard Business Review study showing up to 80% of large global companies have implemented AI into their operational practices, this is a rise from 10% in only 5 years!
Coming closer to home, in the Australian HR context, Sage People surveyed 500 HR leaders in 2020, with 24 % percent revealing they have already implemented AI into their TA approach, while 59% plan to do so across the next 12 months.
The ROI Challenge
“All too often, AI projects start by trying to implement a particular technical approach, and, not surprisingly, front-line managers and employees don’t find it useful, so there’s no real adoption and no ROI”
Roman Stanek, CEO of GoodData.
This is not surprising when viewed through the lens of Deloitte’s 2018 Global Human Capital Trends which found that 72% of people surveyed think adopting Artificial Intelligence (AI) is important for their business, however a minority 31% percent feel ready to address its use and implementation.
In order for HR to help shift the dial on this and before we can integrate our working behaviours with AI successfully, we may need to go back to some simple fundamentals and learn what even is ‘AI’?
Are you one of the 69% who lack clarity on this topic? Yes? Then you're not alone! Despite having worked with HR technology for over 10 years, I too found myself a few years back needing to upskill and improve my knowledge in this area. Whilst acting as a reviewer, purchaser and implementer of HR AI powered tools, I had a clear understanding of the business benefits of the software, but was I fully equipped to review an AI in the same way I did a CRM or ATS? Did I appreciate what was under the hood of this new wave of HR technology? I’ve gone onto study Psychology with a keen interest in exploring how the neural networks of the brain work, and subsequently how these may be mirrored in machine learning algorithms. As a fellow HR practitioner I want to help share the basics of this knowledge with my peers.
Whilst there are certainly pockets of expertise around AI and its implications for use in HR, as seen above the wider HR literacy and maturity around AI understanding is like giving a new toy to a 5 year old where the label advises for use of 10 years+. We should learn to walk before we run.
Validity, privacy, ethics and trust factors are all part of taking on AI powered tools. This mini-series will outline 3 areas HR practitioners who are selecting, implementing and applying artificial intelligence tools should have at least a basic literacy in:
Learn the Lingo: Can you tell your Machine Learning (ML) from your NLP? We’ll kick off this mini-series with some basics explanations of the types AI.
Validity: From Occupational Psychology Theory to AI Adaptations: Are you responsible for selecting or implementing your HR AI powered technology? In part two of this series we’ll take a look at the questions to ask your technology vendors when assessing the validity and reliability of a HR framework when adapted into algorithmic form.
Ethics and Trust How well do you know your HR technology stack? Is it trustworthy? Ethical? For the final part of this series we’ll turn to our role as HR in this context. In the face of near future laws and audit guidelines, what responsibility do we have as HR to help safeguard the use of AI for our employees and candidates?
So what is AI?
Artificial Intelligence (AI) is a broad all-encompassing term, made up of many different types of machine cognition. It draws inspiration from human cognition across modalities from sound and text to imagery. Taking the view from the Harvard Business Review (HBR) Analytics Services, some typical types of AI are:
Machine Learning (ML)
Algorithms that learn from and make predictions based on ‘training’ data it is supplied with. Two main sub-types of ML can be explained by how the data is used below:
Have you ever wondered how Netflix recommends your next binge-inducing show with a high degree of accuracy that you’ll enjoy it? This is Machine Learning at work! Specifically using a probability model called utility matrix (learn more here ) that is processing the likelihood that you’ll enjoy the next romantic series if you enjoyed a bit of Bridgerton (who didn’t!?)
HR Example in Action: Have you used an ATS where the requisition suggests the top 5 candidates who match the role? This is an example of ML in action, whereby the AI training data will usually come from past ‘successful’ hires into your organisation to begin with matched to past job descriptions. Over time the recommendations ideally become stronger the more data it absorbs from both new CV’s, interview feedback, JD’s and recruiter real time input such as why the candidate was the right fit or not.
Natural Language Processing (NLP)
Relates to human-to-machine text and voice understanding. The machine ability to understand meaning in human languages, ‘hearing’ or ‘reading’.
Asking Alexa a question is a prime illustration of NLP in action in everyday life.
HR Example in Action: Chatbots (text or text-to-speech) are a great example of NLP at play, providing engaging real time feedback and answers to common questions throughout the recruiting lifecycle.
Can be thought of as a machines ability to ‘see’ from still images such as pictures, through to an understanding of what is happening in real time in a video format.
HR example in Action: HireVue’s now removed product that reviewed facial analysis of micro-expressions to assign traits and qualities from candidates, as they completed an online video interview. This type of AI technology in HR is somewhat in the growing stages, and as seen with the recent controversy around HireVue’s removal of the technology, there is trust to be built with employees and candidates alike.
Robotic Process Automation (RPA)
Can be seen as the most basic form of AI ( some may even argue it’s not smart enough to be considered AI!). It involves software bots to carry out rules-based and repetitive commands with high degree of accuracy.
HR example in Action: In HR this may be called smart automation and can include optimising your workflows to include as much automation as appropriate for example in onboarding employees or facilitating employee health care plan sign up. A simple illustration of this can be a workflow step change triggering an email sequence to be sent. It may also replace manual action of pulling data from disparate systems into one single source.
Can you identify which of the parts of your HR tech stack uses which type of AI? Or if you are in the process of purchasing a new AI powered HR tool bring your new lingo to the next vendor meeting 😉!
Join us in part two, to learn how to check if those HR theories and frameworks you currently deploy are validated (or not!) in the new age of HR tools.
If you can't wait for part two or you're entering a buying process now, no problem, contact me here.