Sapia labs, our R&D department, has developed a world-first innovation that will help us deepen our understanding of the contextual meaning of words in written job interviews. Called InterviewBERT, this algorithm combines Google’s model for Natural Language Processing (NLP) with our proprietary dataset of more than 330 million words. BERT, meet Smart Interviewer. Together, they’ll usher in a new generation of pre-employment assessment tools and recruitment software solutions.

Put simply, InterviewBERT makes Smart Interviewer, the most sophisticated conversational Ai in the world. Ours is no simple chat-bot – already, Smart Interviewer is capable of discovering personality traits and communication skills, accurately and reliably, using a candidate’s written responses. With InterviewBERT, Smart Interviewer will learn more about candidates than ever before, faster than ever before. With this speed and accuracy also comes reductions to the unfairnesses and biases that plague the hiring process.

Why, and how, are we the first to transform pre-employment assessment technology with BERT?

Through sound Ai infrastructure, we have been able to accumulate a vast and accurate dataset. This dataset grows by the minute – we interview a new candidate every 30 seconds – and, coupled with the expertise of our Sapia labs team, we can assess candidate suitability for a role in milliseconds.

“The smartest companies know that the fairest and most accurate way to assess someone’s suitability for a role is through a structured interview,” our CEO, Barb Hyman, said. “Text increases accuracy and speed of assessing candidates, while removing biases that come through voice or video interviews.”

Dr Buddhi Jayatilleke, Chief Data Scientist and head of Sapia labs, said the team is excited at the finding that InterviewBERT had such a profound impact on trait accuracy.

“Written language encodes personality signals predictive of ‘fit’,” Dr Jayatilleke said. “The ability to understand people through language has limitless applications, and we are excited to keep inventing more ways to use language data for our customers.”

Dr Jayatilleke said decades of research had confirmed that language has long been seen as a source of truth for personality. 

“What our R&D team has proven is just how powerful language data is when you combine it with enormous data volumes and scientific rigour,” he said. “This capability can be used for assessment and for offering personalised career coaching – a game changer for job seekers, universities, and employers.”

Sapia labs will present its findings from a new research paper, Identifying and Mitigating Gender Bias in Structured Interview Responses, at a SIOP symposium in April. 

Artificial Intelligence (mostly Machine Learning) is being used more and more for high-impact decision making, so it is important to ensure these models are used in a fair manner. 

At Sapia, we recognise the impact these technologies have on candidates when used in screening. We are committed to ensuring fairness by making the evaluations more inclusive, valid, unbiased and explainable – this is the essence of our FAIR™ framework.  

The Fair Ai for Recruitment (FAIR™) framework presents a set of measures and guidelines to implement and maintain fairness in Ai-based candidate selection tools. It does not dictate how Ai algorithms must be built, as these are constantly evolving. Instead, it seeks to provide a set of measures that both Ai developers and users can adopt to ensure the resulting system has factored in fairness.

The lack of transparency related to training data and behavioural characteristics of predictive models is a key concern raised when using machine learning based applications. For example, in most instances, there is no documentation around intended/unintended use cases, training data, performance, model behaviour, and bias-testing. 

Recognising this limitation, researchers from Google’s Ethical Artificial Intelligence team and the University of Toronto proposed Model Cards in this research paper. A Model Card is intended to be used as a standard template for reporting important information about a model, helping users make informed decisions around the suitability of the model. The paper outlines typical aspects that should be covered in a Model Card, such as  “how it was built, what assumptions were made during its development, what type of model behaviours could be experienced by different cultural, demographic, or phenotypic population groups, and an evaluation of how well the model performs with respect to those groups.”

Sapia Labs has adopted and customised the concept of a Model Card to communicate a broad range of important information about a model to relevant internal and external stakeholders. It acts as a model specification, and the single source for all model details.

Here are some of the topics covered in a Sapia Model Card:

• Model Details: Provides high-level information about the model under the subsections overview, version, owners, licence, references, model architecture, feature versions, input format, output format. These details clearly set out the responsibility for the model and document all the relevant information.
• Considerations: Important considerations in using this model, such as intended users and use cases, ensuring that the model is used only as originally intended. It also includes a colour-coded summary of adverse impact testing results (covered under quantitative analysis below).
• Dataset: Sources and composition of the dataset and distribution charts of features used by the model.
• Quantitative analysis:
  • Adverse impact testing: Statistics on sensitive attributes and groups, a visual overview of adverse impact testing results in terms of effect sizes and the ratio of recommendation rates (4/5th rule), followed by a very detailed report going into the adverse impact at the individual feature level.
  • Model dynamics: Distribution of the outcome score and the behaviour of the model, presented with partial dependency plots, which improve the explainability of the model.

The generation of the Model Card is automated, and is an integral part of the model build process, ensuring a Model Card is available with every model. 

Having a standardised document for communicating a model specification has enabled faster and more effective decision making around models, especially on whether to go live or not. Integrating Model Cards is part of the continuous improvement process at Sapia Labs on the ethical use of ML/AI. The contents continue to evolve based on the team’s ongoing research and requests made by other stakeholders. As far as we know, this effort is an industry first for the employment assessment industry, and we are proud to be leading in this space.

To find out how to use Recruitment Automation to hire faster, reduce bias and save, we also have a great retail industry eBook on Ai in HR.

AI RECRUITMENT TOOLS

Artificial Intelligence. Machine learning. Chatbots.

While the possibilities of technology always felt like some distant future, there’s no denying that the future is right here and right now.

Every day, technology touches and enhances our lives in ways we rarely even pause to think about. Algorithms, apps and digital automation continue to reshape the ways we shop, connect, bank, get around or even track our fitness and the steps we walk each day.

It’s changed the ways we access customer services and the ways we can connect with our tribes across social platforms. And in the time of COVID-19, it’s enabled ways of efficient remote working that few thought could be possible.

Recruitment and Ai – We’ve only just begun

With the uptake of automation and artificial intelligence (Ai) across every industry sector, it was inevitable that these technologies would reshape the HR and recruitment domains too. Compared to manufacturing automation, service delivery, supply chain management and marketing channels, HR and recruitment might be a little slower on the uptake. Ai tools are now rapidly reshaping the essential functions of hiring.

Ai recruitment tools are used for 3 key functions in the hiring process: sourcing, screening and interviewing of candidates.

Employing the latest advances in Ai, machine learning and big data practices are delivering new efficiencies and better outcomes for businesses, recruiters and candidates alike.

Text and chat interview automation with Ai

Conversational Ai is a type of Ai that lets businesses have dynamic and meaningful conversations at scale with customers, staff, business partners and candidates.

Conversational Ai uses Natural Language Processing (NLP), a sub-field of Ai that’s focused on enabling computers to understand and process human languages. Through machine learning, it aims to get computers closer to a human level of understanding of language.

Conversational Ai uses NLP to discern meaning from both written and spoken word:

• Voice-activated systems ­ – NLP is used in digital assistants you’re probably familiar with: Siri on iPhone, Google Home or Amazon Alexa, for example. These follow instructions to play music, to control connected devices throughout the home, find web-based information and resources and more. On an enterprise-level, you’ll be familiar with voice-driven customer service over the phone.
• Text driven systems – online or on mobiles, chat text discerns meaning in the written word.

How conversational Ai tools are changing the recruitment conversation

Sometimes referred to as chatbots or textbots, Ai-based conversational tools continue to evolve and be applied in new and extraordinary ways.  Through NLP, the ability to read, decipher and understand written and spoken language has evolved to the point that personality traits, sentiment and other inherently human characteristics can be understood from written and conversational exchanges alone. Our own peer-reviewed research shows how personality traits can be accurately inferred from answers to standard interview questions captured via a text chat.

Ai recruitment works best in high volume recruitment such as customer-facing retail or service team roles.  In roles and industries with fewer candidates or more senior positions to fill, traditional recruitment practices are likely to be preferable.

Conversational Ai can be helpful for profiling personalities in candidates or existing employees without the time and costs of conducting lengthy psychometric profiling. Add video into the mix and machine learning can add additional layers of meaning through analysis of facial expressions and profiles, body language and more.

Video interviewing continues to divide opinion as many believe it allows for unconscious (or not so unconscious) bias to remain front and centre of the hiring process. In text-based  Ai interviews, many of the usual bias cues or triggers an be effectively eliminated at the candidate screening stage.

https://sapia.ai/blog/what-texting-language-can-reveal/

Automated interviews support remote working and remote hiring

In a post-COVID or COVID-normal economy, employment opportunities will be competitive. As more people compete for potentially fewer jobs, finding and engaging the best candidates will be even more challenging.

Ai-powered interviews can help recruiters cast their net wider to reach a bigger pool of candidates and find better-qualified candidates.

Mobile-first puts the power in candidates’ hands

People know text and are comfortable with text. So by providing a text chat-based mobile-first experience for candidates, improves the user experience and addresses communication challenges.

Chat-text provides an easier and less confronting interview process for many candidates.

Everyone has a story that’s bigger than their CV and Ai recruitment interviews give every candidate an opportunity to tell theirs. Candidates can choose when and where they complete their interview and standardised interview questions ensure a level playing field for all candidates.

Sapia’s text chat interview automation is blind screening at its best. We’ve removed possible factors that can influence human bias – no CVs, no socials, no videos, no facial recognition and no time limit.  It’s just the candidate and their text answers, providing a fairer and richer experience where candidates feel comfortable just being themselves.

Recommender systems – How Ai supports people making people decisions

One of the most well-known applications of Ai, data science and machine learning is Recommender systems or Recommender engines. It’s how Spotify suggests the track you might like next. Or how Netflix recommends your next binge-worthy series. And how Amazon recommends books or products likely to be of interest.

In hiring, Recommender Systems use predictive modelling to recommend the most-likely best matches of applicants for a role.

Recommender systems guide decision-making by using machine learning to analyse all the data available through the HR lifecycle. From job advertising and clicks, through interviewing and hiring, to employees’ job satisfaction and tenure, data can be analysed to reveal predictive patterns and insights.

Data can find connections that humans don’t, providing valuable insight into what an ideal candidate looks like or where you’re likely to find them.

Predictive intelligence draws a picture of your ideal candidate

Recommender systems can cut through the ‘noise’ by providing a shortlist of top-ranked candidates. This is without burning time, sorting and reviewing potentially hundreds or even thousands of applications. Predictive intelligence shares additional insights on candidates’ values, traits, personality and communication skills. It helps to simplify the selection and guide faster talent decisions.

Machine learning is not infallible.  One important consideration is questioning whether the data being used is not inherently biased. If, for example, machine learning models are built around data from a workforce that historically skewed towards male, the recommendations will inevitably have a male bias. Machine learning should only guide a decision not to make it and, ultimately, it’s always important to have real people making decisions about people.

Interviews – it’s not where you finish, it’s where you start

Reviewing CVs of all candidates can be the most time-consuming part of a recruiter’s job. Especially for large-scale briefs such as retail or customer service teams. In defining a shortlist of potential candidates to proceed to the interview stage it can be hard to differentiate between CVs. It’s also easy to make decisions that may be based on personal biases.

But what if you could start the hiring process with all the benefits of an interview process, without investing your time in them? And what if in the time it would take to properly review just a handful of CVs, thousands of candidates could be screened by interview?

With Ai recruitment tools you can.

Five top ways conversational Ai tools are changing the recruitment game

When it comes to recruiting and hiring, the ability to read the mood as well as the words is a game-changer in candidate assessment. Here are our top five benefits for your business:

1. Reducing time to hire, improving the quality of candidates

Without even having to consider CVs upfront, an upfront screening interview reduces time to hire by providing a shortlist of candidates with the best fit to move forward.

Ai interview automation looks beyond the CV to assess the skills, traits and temperament of candidates. Based on past hires, Ai learns what a successful candidate for your business looks like and joins the dots to find others that match that profile.

Recruiters and hirers can save time reviewing and assessing CVs. With the ability to complete briefs faster, build teams sooner and achieve business metrics, you can be on to the next job sooner. Or free yourself to concentrate on what you do best: building relationships, delivering a better hiring experience or enhancing the onboarding process.

2. Reduce bias & build diversity

Ai-enabled interviewing helps reduce the effects of unconscious bias – the inherently human prejudices, personal preferences, beliefs and world-views that shape our assessment of others. Our biases can easily have a negative impact on candidates and mean you’re potentially missing out on the best candidates for the job. It can also mean employers are missing the opportunity to cultivate workplace diversity and all the benefits it delivers.

Diversity improves employee productivity, retention and happiness. Time and again, research shows that diversity – of background, gender, experience and more – improves employee productivity, tenure and job satisfaction. In 2020, global management consulting company McKinsey confirmed that “The most diverse companies are now more likely than ever to outperform non-diverse companies on profitability”.

3. Cut costs of every hire

Companies that have automated part of their candidate screening and interviewing are not only reaping the benefits of a more streamlined and stress-free process but report an immediate pay-off in time and efficiency savings.

Get your time back quickly and reallocate budgets towards higher-value investments and automation in other areas of recruiting.

Calculate your RoI on interview automation

Use Sapia’s free calculator to:

• Calculate your costs of hiring
• Calculate the costs of your annual turnover
• See the financial benefits of using automation for hiring
• Avoid unnecessary revenue losses

4. Increase the productivity of every recruiter and hirer

Everyone has one part of their job that they could do better if they had more time. Like managing stakeholders. Improving business partnership skills. Or networking to improve talent pools with a focus on those high-end and hard-to-fill roles. Whatever yours might be, interview automation can give you back time to focus on high-value tasks.

Reviewing CVs and managing interviews might not be the biggest challenge in your role, but they are likely to be the most time-consuming. Automate those upfront interviews using the tools and process of Ai recruitment and you can focus on the bigger picture of finding the best fit for every role and meeting every brief with confidence.

5. Give candidates a better experience

While this one’s last on our list of the benefits of Ai interview automation, it could equally be the most important.

Ever since job boards hit the market, recruiters have been inundated with candidate applications. While that’s been good news for potential employers as well as recruiters, it’s not so good for candidates.  Too often candidates make the effort to apply for a position, but then due to the sheer volume of applications they never hear a thing from the recruiter or hirer.  It’s called “ghosting” for obvious reasons.

Ghosting is not just a bad look, it can be bad for business. Candidates can easily share a negative experience on social media. They may also be less inclined to apply again or accept a job offer now or in the future.

With interview automation, you can turn every candidate engagement into an efficient, empowering and enjoyable experience.

About Sapia

Sapia’s award-winning interview automation offers a mobile-first, text chat interview.  At scale, it delivers an engaging and relatable, in-depth interview, followed up with personalised feedback for every candidate. Here’s how Ai automation provides a superior experience to a traditional interview process:

• A familiar and accessible, mobile-first text experience
• No confronting questions or videos
• Candidates can be themselves, completing questions related to role attributes where and when it suits
• Blind-screening at its best with no gender, age or ethnicity revealed
• Candidates are motivated by personalised feedback, insights and coaching tips… and the opportunity to provide their feedback on the process

Find out more about Sapia’s Ai-powered recruitment tool and how we can support your recruitment needs today.

You can try out Sapia’s Chat Interview right now, or leave us your details to get a personalised demo