AI—What's the Buzz?

Digital voice assistants answer the questions we ask, robots at work take away some of the repetitive tasks we face, and chatbots give us extra support when we need it. What's the big deal about AI, and how does it factor into talent assessments?

AI—What's the Buzz?

30 Nov 2018 by  Aon's Assessment Solutions team

The use of Artificial Intelligence (AI) in assessment within organisations is not new. For years, personality questionnaires have been scored and ‘interpreted’ by expert-developed algorithms. But this was just the start. AI in assessment is growing at a rapid pace – and there are questions you’ll be asked in your role as an HR, recruitment, or talent professional.

Do you know the answers?

Question 1: What is AI in assessment?

AI-based assessment is beginning to be part of many candidate and employee psychometric assessments, from realistic chatbot-type conversations with candidates in situational judgement tests to proven algorithm-based decisions made from looking at candidate responses to test questions. The use of AI in assessment now regularly informs HR and talent decision-making.

Question 2: Is AI already being used in talent assessment?

Yes! The use of AI in the workplace has ramped up in recent years, so it’s not surprising that talent assessment too is now being shaped by AI.

The first move to AI in assessment came in the 1990s, when paper-based versions of tests moved to computers, with automated scoring and computer-generated interpretive reports. For the first time, technology was taking on some routine tasks and using algorithms to produce a candidate report. AI is now being used to generate unique test questions on the fly, and in tests that make use of adaptive scoring.

Question 3: What does the AI jargon mean?

Robots in the workplace. Deep learning. Pattern matching. Do they sound like gobbledygook to you? You're not alone. Here are the key terms to know on the use of AI in talent assessment, and how they analyse and interpret vast amounts of candidate data.

Robotic process automation: Gathers and transfers expert knowledge, then programs the system with an ‘if/then’ rule-based approach. Chatbots are a great example, as are computer-generated interpretative reports in talent assessment. But this rule-based system is not capable of learning and improving without being given explicit instructions.

Machine learning: Even though a computer system cannot think for itself, statistical tools can enable the system to model predictions from any given data and improve prediction over time. This is used to create predictive people analytics and help employers make better talent decisions.

Pattern matching: This technique gets the computer to check the sequence of responses to determine if there is a pattern. It can be used to carry out some of the ‘human’ tasks such as recognising faces or identifying emotions.

Natural language processing: Makes use of text and speech analytics to extract the underlying meaning. This could have applications in analysing speech in interview question responses.

Question 4: How does AI improve candidate and employee assessment?

1. Precision. AI can analyse massive amounts of data, much more than any human. Thanks to the increased power of today’s computers as well as algorithms and machine learning, more candidate data can now be precisely evaluated to help make better selection decisions.

2. Efficiency. AI enables recruiters and talent teams to conduct consistent and objective assessments of job-relevant data at a much earlier stage in the selection process. It also enables you to utilise video interviews far earlier in the selection process than is typically the case with in-person interviews.

3. Reducing bias. Humans are vulnerable to biases and stereotypes. This is often why poor selection decisions are made. In theory, AI’s objectivity will help recruiters to eliminate conscious and unconscious bias in the selection process. But, in reality, you have to be very careful about how your AI system is programmed. An algorithm is only as good as the data that’s fed into it. As part of your AI design, makes sure it is not mimicking just one assessor (and all his or her biases), but draws from several assessors.

4. Legally defensible. The AI that’s built into your talent assessment must be transparent and open to challenge. If your selection decisions cannot be easily explained, they could be challenged by applicants in a court of law. Allowing AI to continuously learn by ‘observing’ the best practice of human raters offers the best and most legally-defensible approach for assessment.

5. Engagement. AI can significantly improve the candidate experience in recruitment. It enables recruiters to offer immediate support and help, for example through interactive chatbots that can answer queries about the selection process or about specific assessments. AI can also optimise and enhance the selection experience for candidates. By speeding up decision times, reducing bias, enhancing the assessments and making the process more candidate-centric, AI can improve the whole selection experience for jobseekers.

Question 5: How does AI support video interviewing?

Video interviewing involves candidates being asked to record themselves responding to, typically, competency-based interview questions. These video recordings mean that candidates no longer need to travel for interviews, interviewers get to re-watch and share candidate responses, and less time is spent on the interview itself. But that involves a considerable amount of time on analysing individual responses.

AI can improve the speed and objectivity of this analysis – by transcribing the audio, then analysing it for clarity of speech and proficiency in language. It also helps analyse visual elements through emotion tracking software and facial recognition.

Question 6: What are the challenges of using AI in assessment?

1. Defensibility. Standardised ‘plug-and-play’ AI systems are available but they won’t differentiate your employer brand. If your competitors use the same systems, you’ll all be chasing the same talent. Also, these systems utilise ‘deep learning networks’ which learn as they go. This sounds promising but actually it makes it very difficult to explain exactly why candidates were accepted or rejected. These systems lead you to make selection decisions that you can’t defend, which leaves you vulnerable to litigation from disgruntled candidates. Only custom AI systems offer the ability to make transparent and defensible selection decisions.

2. Time. Custom AI systems mirror human behaviour and replicate the best practice of your assessors and raters. To achieve this, you have to pre-feed the system with relevant information. It can take up to six months to ‘train’ an AI system to assess candidates in exactly the same way that your assessors and raters would judge them. Managing this lead time will be a major challenge for organisations.

CHROs should therefore be forming project teams now to look at custom AI models for video interviewing and other recruitment processes. Otherwise you’ll always be six months behind the pioneering companies that have already invested in this technology.

3. Ethics. There is an ethical question around how much support you take from an AI system. For example, are you happy for an AI system to reject your candidates? Or would you prefer it to ‘flag up’ unsuitable candidates so you can review and check their details? How to use AI ethically will be a key consideration for many employers.

AI’s role should be restricted to providing additional information and enhancing efficiency. Recruiters should always set the objectives when hiring. AI can then deliver useful information, at various stages of the selection process, that will support a final decision.

4. Data handling. AI excels at analysing massive amounts of data. However, when so much data is involved, the results can be misinterpreted or even deliberately abused. Good data handling practices will be essential not just for confidentiality but also for maintaining your organisation’s reputation. AI should be used carefully and honourably to help you predict which candidates will be effective in the role – and engaged by your organisation.

Question 7: How is AI in assessment likely to develop?

Assessment developers are already looking at how AI can help the interpretation and understanding of open-ended questions in personality questionnaires. Real-time interviews could be carried out by an avatar over the internet, or with the avatar being the observer of a hiring manager’s interview. 

Using AI in assessment takes away admin and hassle from HR decision-makers while making sure they are still in control. For candidates, AI in assessment means that they get to respond to assessment tools in a more ‘natural’ way, not restricted to just completing ‘written’ tests.

Request a demo

Need help with integrating AI technology into your assessment processes? Get in touch with us for a demo.

Aon's Assessment Solutions team

Aon’s Assessment Solutions is part of Aon’s global offering in talent solutions, helping clients achieve sustainable growth by driving business performance through people performance. Aon’s Assessment Solutions group, including the cut-e and cocubes brands, undertake 30 million assessments each year in 90 countries and 40 languages.

Get in touch
Ishita Bandyopadhyay
Vishal Singh
Vishal Singh
Mumbai, India