There has been some negative media attention lately surrounding the use of Artificial Intelligence (AI) in the recruitment space with warnings ranging from the fact that AI produces a shallow candidate pool to more serious things like amplification of bias.
There are many instances of AI being used in a way that has harmful outcomes, but it is important to clarify that this is about how AI is being implemented and not an issue with the use of AI itself.
When AI is used appropriately, responsibly, and following regulatory guidelines it is an incredibly powerful tool that can create fair outcomes for candidates who are selected without bias – in a way that no other tool at our disposal can.
This is why we think it’s worthwhile that more people better understand AI and some of the differences in the way it is used and implemented.
Most media articles refer to AI as if it represents a singular master algorithm and fail to identify how varied the implementations of it are. Almost all AI we have today falls into the category of “narrow AI”, in other words algorithms, mostly machine learning, built to solve a specific problem. E.g. classify sentiment, detect spam, label images, parse resumes. These purpose built AI are highly dependent on the nature of the underlying training data and the expertise of the developers in making the right assumptions and tests of validity of their models. When built in the right way and used responsibly, AI has the ability to empower humans. This is why at Sapia.ai we have made various conscious design choices and adhered to a framework called FAIR™ that tests for bias, validity, explainability and inclusivity of our AI based tools.
The biggest cause for alarm is when AI is applied to analysing video, which can lead to irrelevant inputs like clothing, background, and lighting being used as predictors of personality and job-fit. Video and speech patterns also make it nearly impossible to remove demographic information like race and gender as inputs.
Additionally, analyzing facial expressions is problematic, especially when evaluating certain candidates like those with Autism Spectrum Disorder or other forms of neurodiversity.
This is why Sapia.ai does not, and will not, use AI scoring for video interviews or even voice transcriptions from videos or audio given the word error rate introduced in transcribing speech. Instead, we opt for text – which we implement in a friendly no pressure environment that feels like you are texting a friend.
It’s worth noting that no data other than the answers given by the candidate are used in the ‘fit score’ calculation – that is, we never use demographic data, social media, CV or resume data (which also contain demographic signals, even when de-identified), or behavioral metrics such as time to complete.
Even a candidate’s raw text itself contains gender and ethnicity signals that can introduce bias, if not mitigated. This is why we only use feature scores (e.g., personality, behavioral competencies, and communication skills) derived according to a clearly defined rubric in our scoring algorithms, which our extensive research shows contain significantly less gender and ethnicity information than raw text.
Another common concern is that AI will result in more uniformity rather than diversity in the workforce as algorithms narrow the pool in order to search out an employer’s ideal candidate. There are several things worth noting here.
First, identifying what the ideal candidate is – that is, what knowledge, skills, abilities, and other characteristics are important for success in the role – is what a job analysis is for and should, legally, be what your selection tool is designed to measure.
This is also not specific to AI, as all selection systems are designed to identify which candidates have a profile of traits and characteristics that indicate they will likely be successful in the role. This doesn’t automatically mean that every hire is going to be exactly the same, though. When you focus on the traits and characteristics that will set someone up to be successful, considering potential more than background or pedigree, you’re more likely to uncover hidden talent and hire more successful people from a broader, more diverse range.
Relying solely on past data to build your model also runs the risk of introducing historical data biases. This is actually why it is so important to consider the ideal candidate profile and use that to inform your scoring model. We strongly believe in keeping the human in the loop, which is why our scoring models are centred around the human-determined (via job analysis) ideal candidate profile and then optimized to ensure all bias constraints (e.g., 4/5ths rule and effect sizes) are met.
Using this approach, Sapia has helped clients achieve their DEI goals and increase their diversity hires, including impressive statistics like hiring 3x more ethnic minorities, 1.5x more women, and 2x more LGBTQ+ candidates in just 3 months.
Lastly, it’s worth acknowledging that there is often a “black box” mystery of how AI recruitment tools work. People don’t trust what they don’t understand. While we don’t expect everyone to be an expert in AI or Natural Language Processing, we do strongly believe in building trust through transparency and work hard to make sure that our models are easily understood and open to scrutiny. From third-party audits to detailed model cards to in-depth dashboarding and reporting, we aim to maximize transparency, explainability, and fairness.
We believe a fairer future can only be achieved when AI is used responsibly. AI is not the enemy, rather it’s the experience and motivation behind those promoting it that can make the difference between what is good AI and what is harmful AI.
The jewel of Australia’s tech sector, Atlassian, has been lauded for giving staff the privilege of working from home forever. But when I posted the story on our team Slack channel, I added a comment warning of the longer-term impact of “remote forever”.
One of our senior team members replied: “Why do people travel in the morning to an office, in a packed tram/train carrying a laptop, then work on that laptop only to carry it back home in a packed train, wasting precious time? That looked comical to me for a long time.”
When I worked for another technology company, we spent a lot of energy trying to convince leadership that WFH did not mean a free ride and would, in fact, unleash productivity and improve engagement. COVID-19 has brought forward the idea of WFH as an alternative arrangement for many who would not have otherwise considered it.
While we may be revelling in the success of dismantling the long-held bias that you need to see someone at work to trust they are doing the work, it comes with its own challenges around organisational relevance.
Does it matter what company you work for if the only difference between one job and another is for whom you are completing a task, and perhaps the one or two people with whom you work closely?
When we all worked in ofﬁces, some of that intimacy was built by the serendipity of conversations you had while going about your day’s work.
There was always the potential to catch someone from outside your team and share an idea and solicit a different perspective. There was an ease of connections and interactions that can be hard to replicate in a remote work context.
Being remote is a little bit like trying to establish a long-distance relationship which, as many know, has the chances of success stacked against it.
Then there is the inﬂuence of place, and of space. At REA Group, where I worked for some years, the building fed the culture. Its design and redesign had been carefully thought through to maximise connections and space to collaborate – and not just with those in your immediate team.
Why do people go to church to pray, the pub to drink, and the footy to watch their team, when they have the Bible at home, beer in the fridge and a TV in the living room? Because they are looking for connection, community and inspiration.
Once the novelty of WFH wears off, and for many it already has, comes the challenge of maintaining connection, building afﬁliation and building cultures when people and teams are not physically spending time together in a shared space.
Is there a way to assess performance when you can’t see people at work? How do you look out for people, mentor them, develop them, when your interactions are all booked in, bounded within a strict working day? What way to acknowledge someone for something you heard they did well, as you might if you jump in a lift together?
There is a real risk our employment relationship becomes transactional, which affects engagement, which then affects productivity.
We know from our own work in this space, personality is not 16 types on a table. It is way more nuanced and diverse than that. In a population of 85,000, equal men and women, we ﬁnd at least 400 uniquely identiﬁable personality types.
We live in a world of hyper-personalisation, from our morning news feed to our Netﬂix proﬁle based on our viewing history. How can an organisation retain that diversity of perspective. That is when it usually thinks of two binary ways of working: in an ofﬁce or at home? It can’t. That is why the future of work has to involve a new type of technology. One that can navigate the rich mix of types we work with and adapt to their communication and working style.
I have championed for WFH when in senior HR positions. However, this experience highlighted the many things I might have taken for granted in an ofﬁce environment. It has nothing to do with fancy decor and an ergonomic chair. It’s more the human moments of serendipitous connection that disappeared so quickly, almost without time to say goodbye.
It would be great to think we all emerge from this situation with a mind to honour the things we have learnt about our “work selves” and, most importantly, to build company cultures that thrive by accommodating those diverse needs.
To keep up to date on all things “Hiring with Ai” subscribe to our blog!
It’s not every day or every job where you get to say you are changing the way the world works.
For 2+ years, the small team of incredibly dedicated data scientists led by the incredibly humble Buddhi Jayatilleke have tested and re-tested and experimented and re-experimented to find a new formula for assessing talent – one that is 100% inclusive and bias-free, but also human, using the combination of AI, machine learning and advances in NLP.
Apart from reading daily the thousands of comments of gratitude we receive from candidates for this new formula, which is globally unique! it is wonderful to see that team receive the industry acknowledgement at a global level.
Last week, at a Virtual CogX, the world’s largest Festival of AI and Emerging Technology, with top CEOs, Scientists, Technologists, Data Scientists in attendance, with over 30,000+ attendees from hundreds of countries, 6500 world leaders and 650 presenters across 17 forums, this team were awarded Top 3 For Best AI in HR technology.
For a team that has been tackling this problem for such a small amount of time, with limited resources but endless tenacity and commitment, we couldn’t be prouder to get to work with them every day.
The PredictiveHire Data Science Team:
Are you interested in using an award-winning solution in your business to recruit faster, better, fairer? Let’s chat
To find out how to use Recruitment Automation to ‘hire with heart’, we also have a great eBook on recruitment automation with humanity.
From one recruiter to another and one employer to another, the ways candidates are selected vary greatly. But ask anyone involved in the process, and most will agree that what happens at the early candidate screening stage, is critical to getting the best outcomes. Traditionally, it’s also been the most time-consuming and costly part of the hiring process.
Long before a face-to-face interview, recruiters need to screen candidates to decide, from potentially thousands of applicants, who should proceed to the next steps in the hiring cycle.
But before they’ve even met a candidate, can recruiters really assess someone’s ability and suitability for the job they’re applying for? Yes, they can.
In contemporary recruiting, a suite of tools and technologies can help take the hard work and the guesswork out of the hiring process. Talent assessment tools help recruiters identify the best candidates faster – talent who will be the best fit for the role and the team, work most productively and stay in the role longer.
While traditionally a time-consuming manual review of applications and CVs would begin the hiring process, recruiters have embraced technologies that can automate these processes from the outset.
In this article, we compare two top of the funnel tools recruiters are using to assess candidates: traditional situational judgement tests (SJTs) and the next generation text interview platform.
Sapia Ai-enabled automated interviews could provide the answers you’re looking for, helping to connect to the best talent faster and more cost-effectively.
Situational judgement tests are used to assess a candidate’s judgement and ability to respond appropriately to the real-world situations they would be likely to encounter in the workplace.
Candidates are presented with a workplace scenario and then they are required to choose or rank the best (or worst) paths to resolve the challenge, conflict or opportunity. They are a type of psychological aptitude test that provides insight and assessment of a candidate’s job-related skills.
While the challenging scenarios presented to candidates are hypothetical, the best tests are designed around the role they are applying for.
Reflecting real situations they could encounter, the scenarios may involve working with other team members or supervisors, interacting with customers or dealing with day-to-day challenges.
Situational judgement tests date back to the 1940s. While the ways they are delivered may have changed, they remain a popular way to assess skills such as problem-solving and interpersonal skills. They are also useful in assessing soft skills and practical, non-academic intelligence.
Situational judgement tests are customised to the role and the organisation. Generally, they would be looking to assess a candidate’s aptitude for a role by measuring competencies that might include:
As they are produced by a range of different providers, SJTs can be delivered in a number of ways. As they are also tailored to suit specific roles and companies, tests can vary in their length, structure and format. While some may be paper-based ,most tests are delivered digitally.
The tests provide candidates with a workplace scenario – as a written description or as a video or digital animation – and a challenge related to that scenario. Typically, candidates are then presented with four or five possible paths of action in multiple-choice format to deal with the situation described.
Different approaches are used for candidates to provide their answers. Some may require candidates to choose both the most desirable and the least desirable action. Others may ask candidates to choose just one preferred option or rank all actions in terms of effectiveness.
SJTs are typically used before the interview stage and often used in combination with a knowledge-based test.
SJTs are designed to help recruiters and hiring managers to:
Since 2013, Australian recruitment technology specialist Sapia has worked to solve a problem for every recruiter and employer. That is how to get to the right talent faster while consistently improving the candidate experience.
Sapia’s text-based interview platform uses artificial intelligence, machine learning and natural language processing to provide reliable personality insights into every candidate. While SJTs can be expensive time-consuming to create, administer and assess, Sapia’s platform can a like-for-like personality and job-fitness tests with far greater ease and at a fraction of the cost.
Often SJTs don’t accurately represent what the job is really about. There are so many aspects that need to be considered within a real-world situation. Feedback from the SJTs pilot groups is that they often felt as though they were being forced into specific areas that may not be job-related. There needs to be more flexibility for a candidate to say: “I would do this, but I would also do a bit of that”. Having an experience that gives flexibility in answering. It enables candidates to have that open-ended answer to express what was important to them.
Smart Interviewer is Sapia’s machine learning interview platform. With learning from analysing more than 165 million words in text-based interviews with more than 700,000 candidates, Smart Interviewer combines standard interview questions related to past behaviour and situational judgement to reliably assess personality traits. The questions can be customised to the specific role family – sales, retail, call centre, service etc– and specific requirements relating to the employer’s brand and employment values.
The scientific foundation of Sapia’s Ai interview platform is that language forms the framework for the knowledge, skills and personality we possess. Through a simple text-based conversation, Smart Interviewer provides valuable candidate insights. It can predict a candidate’s suitability for a role and guide their progression through the recruitment process. It delivers the insights that recruiters and employers need to make better hiring decisions at scale.
Improving the candidate experience is a priority for every recruiter and employer. The effect of a poor experience can cause lasting damage to reputations and brands. Sapia is the only conversational interview platform with 99% candidate satisfaction feedback. Candidates enjoy the process, appreciate the opportunity and value the personalised feedback. Something that’s simply not practical with most high-volume recruitment briefs.
As text is a familiar, non-confrontational way to connect, candidates enjoy the text interview experience. Unlike SJTs that lock them into choosing options from pre-determined answers, candidates appreciate the open-ended questions . Here they are empowered by the opportunity to tell their story in their words.
While questions are customised to the role, some typical examples include:
• What motivates you? What are you passionate about?
• Not everyone agrees all the time. Have you had a peer, teammate or friend disagree with you? What did you do?
• Give an example of a time you have gone over and above to achieve something. Why was it important for you to achieve this?
• Sometimes things don’t always go to plan. Describe a time when you failed to meet a deadline or personal commitment. What did you do? How did that make you feel?
• In sales, thinking fast is critical. What qualifies you for this? Provide an example.
Sapia provides blind-screening at its best, effectively reducing opportunities for bias from the assessment process to ensure every candidate is playing on a level field. Candidates recognise and appreciate the opportunity to tell their story without the subjective biases of a human interview or a cursory review of their CV. For top of the recruitment funnel interviews, Sapia removes CVs from the process altogether.