With AI as such a new technology, you’re navigating new territory. We know there are challenges in implementing AI simply because it’s new. Fortunately, together with our customers, we’ve found great ways to tackle these common concerns and compiled them in HireVue’s AI for Hiring: The Buying Guide.
We also asked the analyst community to share their thoughts about navigating the AI-driven HR technology space. These are their insights on finding the right partner:
Stacia Sherman Garr
Co-founder & Principal Analyst, RedThread Research, @StaciaGarr
HR projects often have a lot of moving parts, and frequently have unintended consequences that extend beyond the initial scope, particularly in regard to diversity and inclusion. HR professionals should make every effort to understand the full benefits and limitations of solutions they evaluate – including those downstream effects that may not be immediately apparent.
A critical question to ask technology vendors is if they conduct algorithmic audits and risk assessments to determine how their predictive tools reach decisions and their potential impact on diverse people.
Read more about algorithmic audits in How Humans + AI Overcome Hiring Bias.
Principal Analyst, Lighthouse Research & Advisory, @BenEubanks
AI has the capability to reduce bias and minimize some of the common human characteristics that jeopardize hiring decisions. However, it is not infallible.
Employers should not be afraid to ask vendors how decisions are made and what signals are taken into account when making recommendations to ensure that they can find the best ways to blend human creativity, critical thinking, and curiosity with the best that machines have to offer.
Read more about bias reduction in: The Impact Of First Impressions On Hiring: A Data Science Perspective.
President, HRU Technical Resources, @TimSackett
Buying AI for recruiting is like buying a computer. You have no idea what’s on the inside, and how it works. Basically, you just want it to work! The key then to buying the right AI tool for you is all about talking to someone who is currently using it and gaining a better understanding of how they are using it, what they like about it, what they don’t like about it, and what impact it’s had on their metrics. Just because it works in demo means nothing in it working in real life.
See how Unilever finds better candidates faster with AI in their case study.
Principal Analyst Talent Acquisition, Brandon Hall Group, @DariaFriedman6
While HR practitioners should always consider systems support, better results, and meeting current and future needs when purchasing any technology, important to AI are the questions related to learning over time, providing output explanation, enabling some user feedback, and mitigating unconscious bias.
With respect to learning over time, AI-driven technology solutions can base output (scores, etc.) on historical data, ongoing data, and user feedback. With some solutions, the learning takes place behind the scenes, in a “black box”, while other AI-driven solutions provide some information that explains the output, such as level of experience in position for a candidate selection. The point is if the solution learns over time, the organization is more likely to see improved results over time. If the solution provides some explanation for the output and the user can provide some feedback, like a “thumbs up or down,” then the organization may feel more confident about the decision to be made.
See How Artificial Intelligence is Transforming Pre-Hire Assessments for a deeper look into AI in recruiting.
Research Analyst, Nucleus Research, @TrevorW50701401
HR customers should focus on vendors that are committed to usability today. Usability should outweigh functionality as HR staff move to AI software for the first time.
For a longer-term view, organizations should make sure that their vendor is focused on growing AI capabilities in the future to keep pace with market advancements. Lastly, buyers need to make sure that they understand any underlying bias in the algorithms being used to make sure that they are complying with anti-bias regulations.
Read more about moving recruiting teams from transactional to strategic with AI technology in the Empowered Recruiter eBook.
Director of Research, Fosway Group, @DavidPerring
Recruiting is probably in its most competitive phase ever. But, whilst many organizations struggle to create a recruiting experience that works for candidates, existing employees, managers and for HR; a growing number are beginning to get a serious lead in their approach.
Smart organizations are realizing, that often, it’s not their core platforms that give them the advantage. If you want to attract and retain the best, it’s the intellectual property that’s also brought to the table, on top of the platform, that frequently makes the biggest difference. Harnessing AI and embedding intelligence into the experience of everyone who’s involved in the hiring process is creating a step change in the arms race for good people.
Read more new ways companies are seeking talent advantages in: 5 Ways Recruiting Is Poised To Evolve In 2019.
Planning for, buying, and implementing AI can be very effective if you are thoughtful about the problems you need to solve and the processes you need to improve.
Just because everyone else is looking at buying something – or a large number of vendors suddenly crop up with very similar products – doesn’t mean it will address the critical issues of your company. Keep your eye on what will improve your processes, but also the outcomes, like the quality of candidate that you hire.
Navigating the market doesn’t have to be confusing.