AI & unconscious bias in recruitment:

A variety of AI tools are used in recruitment marketing and digital recruitment. There are AI tools for every step of the recruitment process.
One of the most valuable applications of AI is the ability to post improved targeted job ads that will be seen by the right candidates at the right moment. AI solutions like Textio help recruiters improve their job postings, making it easier to attract and differentiate candidates. Recruiting chatbots can add more information and have a basic conversation with candidates. Advanced chatbots can even analyze their responses to see if a person demonstrates certain skills. A variety of resume scanners are offered that screen and surface promising candidates. There are also vendors that offer AI-powered psychological and gamified assessments and automated interviews.

Benefits of using AI in recruitment marketing:

Better sourcing and screening

AI tools can help create a wider and better targeted candidate pool. Recruiters can quickly collect more data on each candidate and evaluate them more effectively.

More actionable insights

Through predictive analytics, AI tools offer insights into job seekers’ behavior that recruiters can use to design their strategies, define more realistic goals, efficiently track their efforts and make more effective decisions.

Saves time and money

AI powered tools analyze big amounts of data in a few seconds and provide understandable results that can be considered by decision makers.  By saving time, AI solutions also save money and resources.

Eliminate human bias

Biased decisions are a huge problem in recruiting and a major barrier in efforts to improve diversity and inclusion in the workplace. Whilst training and awareness can help, as humans we have inherent unconscious biases. With proper programming, AI solutions are able to minimize this element and improve diversity. This covers anything from racial and gender diversity to age and even geographical location.

What unconscious bias means in recruitment?

Unconscious or implicit biases are learned stereotypes that are automatic, unintentional, deeply engrained, universal, and able to influence our behavior. Unconscious bias occurs when our brain makes snap judgements on situations or people without us realising. In recruitment, unconscious bias happens when an opinion about candidates are formed based solely on first impressions. They are automatic and occur without our awareness. Unconscious biases happen even when recruiters try hard not to discriminate against a candidate on the basis of their gender, race, socioeconomic background, sexual orientation, religion or any other factor.

So how can technology help? And what steps recruiters should take to minimise bias through the use of AI technologies ?

How can AI reduce unconscious bias?

AI technologies reduce unconscious bias in two ways:

1. Sourcing and screening decisions are based on data points. AI software sources and screens candidates by using large quantities of data. AI assesses these data points objectively – free from the assumptions, biases, and mental fatigue. Data points are then combined using algorithms to make predictions about who will be the best candidates.The results can be tested and validated. 
2. AI can be programmed to ignore demographic information about candidates

AI tools can be programmed to ignore demographic information about candidates such as gender, race, and age. It can even be programmed to ignore other details such as the names of schools attended and zip codes that can correlate with demographic-related information such as race and socioeconomic status.  
AI technologies can help to hire only the best candidates based on their real potential and personality, as all the decisions are made taking into account data and resume only.

Of course, it’s not that simple. AI algorithms are not inherently neutral as they are trained by humans. 

AI is trained to find patterns in previous data which means that any human bias that may already be in your recruiting process can be learned by AI. A machine learning algorithm is only as good as the quality of the information that goes into it. Human oversight is necessary to ensure the AI isn’t replicating existing biases or introducing new ones based on the data we give it. AI application in recruiting still requires a human oversight to eliminate unconscious bias

• Remember that AI is as only as good as it’s design - who is designing the programmes. Is there an unconscious bias built in? 
• Recruitment data is self-generated - it is not financial data that is fixed, or independent, by definition a CV is user generated and as such is liable to be ‘gamed’. AI tools are therefore limited by the data that is inputted into the system. 

It is not possible to correct human bias, but it is possible to identify and correct bias in AI as recruiting AI tools can be tested for bias. If AI does show a bias in your recruiting, you can act on it. Use your human judgment and expertise to decide how to overcome biases and improve your recruiting process.
When applied correctly, AI can definitely be a powerful tool to help recruiters deal with unconscious bias and improve diversity.
To ensure AI eliminates human bias, HR leaders should:
Assess the nature of their organisation’s diversity issues to determine their goals and point the AI application at the right problem
Look for evidence that a particular approach has successfully helped companies achieve its diversity goals
Look for conscientious vendors that continuously share the validation results for their algorithms and select technology that has been properly tested and validated
Consider training, certification and other process improvements beyond technology in order to create best practices that support diversity in hiring