Working in HR and AI, one of the first questions that often crops up early on in discussions is a nervousness about ‘gathering more data’. The new stringent GDPR legal framework has, understandably – and rightly – made many companies more careful about how and why they gather and process candidate data as they recruit. So now might seem an unwise time to add another level of data analysis and gathering to recruitment: we may be more inclined to advocate pen and paper over a sophisticated online system. But we’d be wrong.
A lean, clean, unbiased machine
Far from adding legal complexity, using AI for recruitment trims out unnecessary data, making it leaner, smarter and less biased — and reducing the burden of data management and processing at the same time. It’s a pleasing irony that the ‘diversity’ data we collect for AI systems is only used to check and report on the lack of bias demonstrated by these systems.
Our clients are sometimes a bit jittery about ‘collecting all that data’ — but we’re convinced that it’s a win-win. Our concept of what data is, and the risks associated with managing and using it, needs updating in line with just how smart data science can be in application.
Recruitment trends in tech for 2019: AI and predictive analytics
Major changes are occurring in the ways human resources and other related professionals find the right people for open positions. Kayla Matthews looks at recruitment trends in tech. Read here
What about all that data?
One of the most common misconceptions about AI is that it has an insatiable appetite for data. Yes, clever systems are often capable of doing clever things with lots of data. But it’s not a requirement. There are a few aspects to that particular data myth that it would be useful to bust.
First, predictive people analytics doesn’t need, and doesn’t do, bulk data collection. AI systems in HR just don’t need the vast majority of information that HR systems and processes collect. So neither bulk collection nor system capacity are issues.
Second, the data that AI needs must be ‘blind’. Then it’s also processed ‘blind’. The data requirements are extremely focused. This intelligent, focused, necessary data ticks all the boxes in terms of GDPR, and future-proofs HR for future similar legislation. In fact, far from complicating data arrangements, GDPR is easier with analytics.
Businesses struggle to make the most of data analytics due to skills shortage
Innovating new business models and maximising revenue and profits should be the next set of priorities for data analytics, according to Infosys. Read here
What about people’s data privacy?
With AI systems, it’s easy to manage how much data you want to share, and how — both for organisations and, crucially, for candidates. An independent platform for recruitment means that candidate information isn’t necessarily brought in-house. As a candidate, you can choose whether or not to release the whole suite of your information to the company assessing you. And, of course, if you want to use your assessment information for your own personal or professional development, you can. At Cognisess, we find many clients receive some of their best feedback from unsuccessful candidates who enjoy, and make good use of, their own results.
Perks of the job
As it goes about its main business, AI will also go far in helping to clean and smarten company data. How many businesses or organisations have perfectly up-to-date data? And of these, how many have a complete suite of HR and business data systems that are seamlessly integrated? One sophisticated AI platform for HR will integrate data from across these and pull them into one system. It’s a secondary benefit, but it’s not one to underestimate.
AI: the new frontier of the HR space
Artificial intelligence (AI) is the next big frontier and, although it’s too soon to predict its full potential, it is already transforming the way we work. Despite job loss fears, the reality is a different story. Read here
Trust issues explained
New legislation will continue to introduce increasingly stringent checks and balances for data, and businesses would do well to get ahead of the curve in terms of compliance. And in doing so, we should bear in mind that the levels of trust and understanding of AI will not always be so low. With AI already in the public domain — finance, investment, medicine, shopping, music, customer service and more, the notion of ‘black box’ AI, and its mythical unseen powers, is fading fast.
Forward-thinking businesses will anticipate this when deciding to adopt AI in their central processes, rather than risk being left behind.