In this comprehensive guide, we will look into every aspect of artificial intelligence (AI) recruiting. Not just the benefits, but also the negative aspects, what you should consider, and how unconscious bias can affect your implementation. We aim to give you a nuanced/thorough picture of how AI recruiting works and why you as a hiring manager should consider implementing it.

So let's dig in.

‍

In recent years, artificial intelligence software solutions have seen more and more adoption in a number of different areas. Businesses save hundreds of millions of dollars annually by deploying machine learning onto all sorts of tasks. With the entrance of ChatGPT, the technloly has taken major advancements and is able to help a significantly larger user group.

With many repetitive, time-consuming, and data-driven tasks, the recruiting sector is a business area with enormous potential for an AI-revolution.

The last major tech transformation within Human Resources was the entrance of applicant tracking systems (ATS). That meant a great boost of productivity for recruiting teams, but many believe that AI recruiting and access to Large Language Models (LLMs) will multiply outcomes many times over.

AI recruiting is a subset of recruiting automation and aims at the exact same thing —  streamlining your recruiting efforts. But as traditional automation has limitations in terms of understanding natural language, analysing large amounts of complex, unstructured data and finding patterns on its own, that’s where AI excels. Even relatively complex tasks can be automated on the basis of taking in huge amounts of data and extracting real value, as long as there is enough training data to feed the system.

Generally speaking, all tools that in some way help make recruiting more efficient, and base their automation on ever-improving machine learning models, fit under the umbrella that is AI recruiting software. From something as simple as understanding the best time of the day to send out an outreach email to a potential candidate to futuristic AI-driven interviews. The limits of what’s possible for machines and AI can do in HR have just been moved forward — by a lot.

Experts in the field anticipate AI to change the very foundation of recruitment and HR. As a recruiter, business leader, or HR manager, keeping up with this trend is vital to stay competitive in an ever-increasing arms race for talent.

Keep reading to get a deeper understanding and see what artificial intelligence can achieve in recruiting.

‍

A number of tasks among the recruiter’s responsibilities are already automated in order to boost efficiency. With the next generation of automation tools, powered by AI, the goal is still saving time and money, while optimizing results. But many tasks can’t be solved using traditional, binary automation tools.

”If you think about the smartest, most switched-on person you've ever worked with, and then think about the biggest slacker and do-nothing person you've ever worked alongside, the contrast between those two people is obvious. Yet no ATS in the world could distinguish between them, as long as the two people worked at the same job in the same company at the same time. Applicant tracking systems don't inquire about what you learned at a job, what you left in your wake or what you view as your greatest accomplishment.”

‍Liz Ryan in Forbes

‍

It might sound like a paradox, but introducing AI can actually help humanize the recruitment process. This new technology can be applied to tasks with a greater level of complexity and can personalise processes at great scale while increasing support to human workers.

Additionally, and adding to the economic incentives, many vendors aim at increasing the results of their recruiting efforts, such as predicting future performance of candidates or reducing bias in the selection process.

By outsourcing the initial screening to AI, human prejudice gets detached from the initial screening. Unconscious bias in recruiting can lead to a variety of issues and the longer you can keep the selection process objectively, the farther candidates can be correctly assessed based on their skills and experiences. AI tools are coming to help, but have to be trained accordingly to not introduce human bias into their deep learning models.

Most importantly, however, is to support overloaded recruiters and enable them to use their knowledge and experience on tasks that require a human touch and expertise while cutting time-to-fill.

‍

In a recent survey administered by Oracle, only about 10% of companies claim they are using AI extensively throughout the hiring process today. While 36% state that they expect a high increase in usage in the coming years.

Naturally, the shift towards powering recruitment by artificial intelligence started where the needs are highest. Established high volume hirers and industries with a underserved candidate pool have traditionally spent endless resources on recruiting and now lead the charge toward a change.

But, as performance continues to increase and use cases expand, it’s expected that many more will follow.

‍

‍

Recruitment automation is nothing new and has been evolving for many years. The introduction of the Internet had far-reaching effects for the recruiting community and caused a major upsurge in the pace of technology adoption.
Here are just some of the examples of how a less constricted labor market, and increased internet adoption and tech support has changed the recruiting sector:

• Bulletin boards & newspapers → Internet job boards
• Excel sheets & custom solutions → Applicant tracking systems
• Job fairs & in-person relationship building → Linkedin
• Overserved labor demand → Employer branding becomes essential
• Non-intrusive recruiting → Social recruiting for passive candidates

In the past years, technology has been increasingly refined and new functionality such as knock-out questions, video interview platforms, and resume databases with keyword matching, is making the life of recruiters’ much easier.

The problem with computer-aided recruiting has always been the computer’s inability to understand the human language. And since recruiting is very much an area where humans are in focus, there’s a mismatch between binary-thinking machines and the philosophical and colorful minds of humans.

That’s where AI comes into the picture.

As an example, in the case of knock-out questions. They are widely used in the recruitment process and eliminate all applicants that don’t meet the requirements. It’s surely a great way of not wasting any time on a job seeker that aren’t qualified anyway, right?

Well, let’s say a knock-out question asks for:

Do you have 5 years of management experience?

Yes → Great, you are still part of the selection.
No → Sorry, you’re not what we’re looking for.

A very binary outcome without any room for flexibility, which raises a couple of questions.
How many great candidates with 4 years of management experience and extremely high ambitions are lost? Is someone without passion for their work but with 10 years of managerial experience a better choice by default?

A smarter solution could be an AI-driven screening interview where grit, passion, and motivation constitute an important part of the overall judgment of each potential candidate.

‍

And with LLMs such as ChatGPT, there is bound to be countless new groundbreaking tech innovations paving the way for big changes on the HR tech scene.

‍

Keep reading for more examples of how artificial intelligence simplifies hiring.

Virtually all solutions are built to save time and money and/or increase quality of hire. The end result may vary substantially.

Let's walk through some of the most interesting ways recruiters are using AI in their daily work.

Sourcing

Sourcing candidates is one of the hardest and most time-demanding aspects of the recruiter's job. Today, finding and attracting new talent is achieved through different tactics. Organically, through building a great employer brand, or by receiving recommendations from current or past employees. Or by sourcing passive candidates through cold-emailing, calling, or putting out a paid job ad.

This is an area where many recruiters spend a lot of time trying to innovate, find new methods, and reduce costs. And some tactics are pretty clever. Such as posting job ads in the HTML-source code where only developers will look.

AI can help with sourcing in a number of ways. For example:

• Parse and search LinkedIn, Facebook and many other platforms to find qualified candidates with the right background.
• Sending personalized outreach emails.
• Easier rank matching candidates (hard with boolean search).
• Predict the likelihood of a candidate to make a move.

Screening

According to Gartner, recruiters spend approximately one-quarter of their time on candidate screening activities, such as reading through a big pile of resumes.

As the application processes move further and further towards one-click experiences, the pile never seems to stop increasing. Recruiters certainly have many other tasks beyond reading resumes, but the vast amount often causes an endless wait for candidates and a lasting bad impression.

This is where AI-driven tools are coming to the rescue.

On one hand, casting a wide net is always a good start when looking for a job candidate, but the quality and actual interest from many applicants can be very low when applying only takes 30 seconds. And getting to the candidates that are actually qualified, interested in the position on offer, and have all the prerequisites in place is like looking for hay in a stack of needles.

Here’s a list of tools that can relieve the burden in screening:

AI-driven resume parsing and semantic matching

Parsing is nothing new, but the understanding of extracted information has had a major upswing through advancing Natural Language Processing (NLP) technology. As resumes have no set standard and there are over a hundred different ways of just writing the date, rule-based parsing doesn’t really work.

For example, the term Davies could refer to either the name of the applicant, the Davies college or the company Davies Group. Context analysis is needed to parse the information correctly.

NLP-based resume parsers have been proven to achieve up to 87% accuracy which is coming pretty close to the average human accuracy of 96%.

AI resume parsing —  main benefits:

• Eliminates the need for either candidates or recruiters to re-enter information into set forms (better user experience and less manual work)
• Reduces the risk of interesting candidates falling through the cracks due to resume reading fatigue.

What to watch out for and keep in mind:

• There are many different takes on parsing resumes, be sure to do some research before choosing a particular vendor.
• NLP is under heavy development and not perfect. There is always a chance that eligible candidates are sorted out due to writing style, not mentioning certain keywords and other misconfigurations.
• Language support beyond English can be limited

AI-driven interviews

AI-driven interviews refer to either a video interview that is assessed by AI or an interview directly facilitated by a conversational AI.

AI assessments
In AI-assessed interviews, the candidates are asked to respond to a set of questions while filming themselves. The interviews are then analyzed using an AI recruitment tool and the candidates are automatically ranked before any human interaction takes place.

The norm is to train a model on existing employees and then use techniques such as face recognition and language/content analysis to predict future performance of interviewed candidates.

These systems have caused quite a hype, especially in the US where they are both being celebrated as a major time saver, but also extremely criticized for not being transparent enough and introducing additional bias.

‍

‍

AI-facilitated interviews
In AI-facilitated interviews, the process is directly supervised and orchestrated by an artificial recruiter built specifically to screen candidates and is oftentimes augmented by LLMs such as ChatGPT. The most common application is through chat messages, where candidates are dynamically asked to respond to questions of different nature. This is oftentimes not the first step of the process, and the interview can involve checking the candidate's attitude towards uncomfortable working hours, or follow-up irregularities from an automatically parsed and analyzed resume.

AI interviews — main benefits

• Sort out any perplexities before entering the manual process
• Speeding up time-to-fill
• Can reduce unconscious bias

What to look out for

• Can lead to increased bias due to questionable AI-models
• AI assessments require training from many existing employees

Cognitive and Personality Testing

Testing has been common in certain trades for quite some years, for example in developing or higher management, tests are prevalent. But for other professions, testing has traditionally been pretty scarce.

With AI entering the pre-employment evaluations, testing can be performed at a much greater scale and with increasingly reliable results. By testing your entire team you can find out where you might have a gap, either domain-wise or personality-wise. Use this information when hiring to build a more diverse team.

AI-testing —  main benefits

• Find out what personalities your team is missing and fill the gap
• Predict future work performance

What to look out for

• Conducting test with a high confidence level takes time
• Make sure your provider has a good way of making sure the right person is actually taking the test.

Reducing bias in the selection

Provided you are using an AI-based recruiting system that is bias-neutral, you can focus human interaction towards the late stages in the recruitment and therefore reduce the impact of unconscious bias early on in the process.

The problem is just that it is tough to construct a system that is free of human bias to start with. Bias can enter through either the data that is used to train the system or by the constraints you set up when designing it.

If unconscious bias is a major pain point for you, make sure to thoroughly examine how your provider handles these issues.

Automated shortlisting

Automated shortlisting is a combination of early and middle recruitment funnel screening methods, for example, resume screening and AI-driven interviews in combination.

The idea is to let an AI application handle the entire process and in the end, present a human recruiter with a few interesting candidates for you to look into.

You should be wary of new regulations that have started to come into place surrounding AI recruiting. The change mainly concerns system transparency. Vendors must be able to show the logic behind decisions to both the employee and employer in order to offer insight into the process and to understand where improvement is possible.

Read more about the US policy on AI in the recruitment field here, and in the EU here.

AI-shortlisting —  benefits

• A major timesaver
• Reducing human interaction and therefore unconscious bias

What to look out for

• Compliance with new regulations
• How are the ranking factors weighted?

Candidate Database Management

Having an updated database with candidate profiles can save huge amounts of time when its time to add a new head to the team. They might not have been the perfect fit for the last opening but might be just right this time around.

Before contacting the candidate it would be good to know about his/her current working situation, what new competencies he/she might have gained, any new certificates or courses completed, right?

AI makes it possible to automate the entire database management and keep it updated and tidy by continuously checking for outdated or lacking information and then automatically reaching out to the right candidate to fill the blank spots. Depending on how much data the AI systems can attain on each candidate, it’s even possible to predict when someone desires a move and then correlate that information with available positions.

DB-management with AI —  benefits

• Keeping your database in perfect shape
• Be able to closely match the supply to the demand

What to keep in mind

• Applicants don’t want to be bothered all the time without any clear cause besides updating your database. Don’t pester candidates too often to keep your employer brand from taking a hit.
• Make sure your talent pool management is as smart as possible and exploit the possibilities that exist in terms of behavior prediction.

Social media analysis of candidates

People show who they really are on social media. Partying 5 days a week might be an easy sell on Instagram but maybe not the type of behavior that correlates with a high performing employee. And what about other, even more, questionable behaviors? Hate speech, cyberbullying, threats of violence, and obscene language? Truth is that social media analysis before hiring is already common practice, just not officially in every sense. AI makes it possible to assess high volumes of applicants and conduct thorough analysis quickly and without recognition bias that might take place if a candidate went to the same school as you.

Social media analysis — benefits

• Get a more realistic image of the candidate and form a better understanding of who they are.
• Mitigate the risks of social bias related to social media analysis

What to keep in mind

• Legal issues. Be sure to check thoroughly what regulations apply in your state/country.
• Not everyone is using social media and that doesn’t mean that they’re not great candidates. Others have accounts but rarely make any updates and lastly, some have fake accounts.

‍

‍

Using AI in the right way throughout your hiring process brings big benefits over the entire breadth. Bad hires cost companies hundreds of millions of dollars each year, and improving the end results can support long term success for you and your company.

All companies could benefit from using an appropriate AI-setup but there are some applications where it makes even more sense:

High volume recruiting

Are you looking to hire hundreds or even thousands each year? When looking for low to medium-skilled workers at this scale you are bound to have endless amounts of applications pouring in. In fact, Glassdoor has shown that corporate jobs attract 250 applications per opening on average. Many of which are not qualified for the job in any way.

Time and cost issues/ downscaled recruiting team

HR and the hiring team unquestionably fill an incredibly important function, but in many companies, it is still seen as a supporting function to the core business. This is why the human resource department is often underfunded with far too many problems to solve in a far too optimistic time-frame.

Outsourcing some of the most time-intensive jobs to automated systems means that HR can do more with their time and keep the spending low.

‍

Suddenly, the world's united hate towards chatbots has swiftly turned into cheerful amazement at how useful chat technology has become.

‍

ChatGPT and similar LLMs hold a lot of handy tricks up their sleeves and can accomplish most text-related tasks you throw at them with decent results. For recruiters, this means, for example, generating drafts of job ads, candidate communication messages, outreach emails, generating relevant interview questions, summarizing resumes, and many other tasks that are needed on any given day of recruiting.

‍

For candidates, it also opens interesting options. For example, generating instant resumes with a good chance of being shortlisted or passing a coding test for a $183k per year salary job for you.

The first iteration of most texts is often a bit bland, but using your imagination to further hone your request can make the outcome surprisingly good. Following up with additional prompts can deliver the exact right kind of vibe. But its capabilities don’t stop at generating text. It can also analyze and simplify information for you.

For example, if you have collected lots of open-ended responses from an employee feedback survey, you can use ChatGPT to identify the most recurring feedback in a much better way than most survey tools can.

Other use cases include devising business plans, writing poetry, generating ideas, and providing sentiment analysis for example.

Make no mistake about it. Large language models and generative AI like ChatGPT and its future iterations will have far-reaching effects on society, employment, and human job descriptions. How we work will most likely change drastically as more sophisticated AI models pop up and new use cases keep appearing.

Humans that can leverage ChatGPT to be more efficient have lots to win. In fact, cooperating successfully with these kinds of models could become a necessity rather than a sought-after skill faster than expected.

‍

As there are many different kinds of applications there is no general way of describing how the systems work. One major differentiating factor is how the AI models are being trained. They can be pre-trained on external data, custom trained on your own data, or a combination of both.

All options above come with their own set of pros and cons.

Pre-trained systems are operating on the basis of historical data from the vendors’ past projects. This is great as long the data is relevant to your use-case. If the system has been trained on screening 10 million bankers for example, and you want to use it to screen a grocery store worker it comes with a risk of error. However, if you want to use the system to screen bankers for a different bank, a large set of banker-specific training data will be invaluable even if the positions aren’t exactly the same.

The other option, which is common in AI-assessment of video interviews, is to train the AI models on your existing employees. This means testing every employee in the team and give the system input such as characteristics of high performing employees. By doing this you will build your own custom model from your workforce which gives less room for errors related to mismatching data. You will also get an overview of what competencies/personalities could be missing in your workforce. But, for machine learning to do its thing, you would need a rather large set of employees, preferably a couple of thousands.

‍

The lasting impression of the hiring process is a hugely important factor in the recruiter game. It’s been shown to impact brand perception, sales, profitability, and many other important areas.

Many claim that extensive use of AI throughout the recruitment process can lead to a static assembly line type of recruitment without any attention to human emotions.
But most providers know how important the candidate engagement and experience is, and actually add large improvements to the overall perception. It’s not just the recruiters who are suffering from an overloaded recruitment process, candidates have awful experiences as they take their time to craft and perfect their resumes, sending it in with high expectations and then don’t hear back for several weeks. If ever.

Certainly, an automated interview is not as pleasant as the real thing, but a typical recruitment process involves several hundred applicants. The realistic option for the vast majority today is no interview at all.

”The possibility of interviewing all candidates that apply is not only helping us build a very detailed database, it also gives the candidates the opportunity to stand out and express themselves in a whole new way. And we save a ton of time by not having to call as many candidates.”

- Cassandra Mastenstrand, foodora Recruiter

Even if the feedback that candidates receive is computer-generated messages, it's often much better than not receiving anything at all. The notion of using AI in recruitment leads to a robotic process is beginning to wash away.

”AI is one of the core smart solutions that can, in combination with richer data sets and better employee and manager insights, really augment to better and faster decisions and provide employees and managers with better, more integrated, and personalized experiences to meet their long time needs.”‍

- Nico Orie, VP People & Culture Coca-Cola

Adding to the time-aspect, candidates that are scheduled for an AI-driven job interview can choose for themselves both when and where the interview takes place, putting them in charge of the next step.

‍

Recruiting is, and has always been, an area with the human in the center. Naturally, as with most technology revolutions, there is resistance in outsourcing human work to machines.

It certainly is true that machines lack many qualities that we humans use to a wide extent in recruiting. Empathy, creativity, and emotion are all important when finding the ultimate candidate for a job and can hardly be replaced with software.

And you can rest assure that a world where hiring is fully automated is far away still. AI programs are designed to aid human recruiters in supplying frameworks for easier decision making and eliminating repetitive tasks.

Still, there are a couple of things that are holding companies back from further adoption.

1. Bias

Most AI tool vendors claim they reduce bias, or even completely remove it. But in the history of AI recruitment, there are some pretty iconic examples of when bias was actually amplified instead of reduced.

Our advice is to be cautious towards any vendors who claim that they are removing bias from recruitment. Make sure they have solid responses to the following questions:

• How are their systems trained? Own data? Bought data?
• What kind of data are their systems trained on? What sectors, what specific positions, what degree of diversity, is the data diverse and representative?
• How do they systematically evaluate the training data to exclude bias?
• Are their systems compliant with the suggested policies in both the EU and the US?
• Are the systems fully transparent and all decisions traceable?

The first step towards reducing bias is to be aware that it exists. Practically any set of training data has some sort of bias, there’s no getting away from it. It’s human nature to have biases that help us make decisions, but in recruiting, these biases can sometimes do more harm than good.

What’s needed is a watertight self-evaluation test that reliably ensures that the software can actually make decisions based on factual patterns and not strange coincidences.

2. Too busy to streamline processes

The HR-department is undoubtedly incredibly important to any company, but none the less often serves as a supporting function to the core business.

And as you are probably well aware, this means that recruiters and other HR-functions are often put under immense pressure to make ends meet and to perform with very limited resources.

AI has the potential to make life easier for most recruiters and HR departments, but finding time for implementation and disrupting processes is for many HR professionals just never a good time.

3. Knowledge and understanding of how it works

One of the most common objections to using AI technology when recruiting is that “AI is dehumanizing the recruitment process”. But as you can find further up in the guide, the candidate experience is often at the very center of most AI recruiting tools. What’s missing is often efficient communication from the vendors combined with a limited understanding of the platforms.

According to Oracle, only 12% feel strongly that they have knowledge about the topic of using artificial intelligence (AI) for the purposes of enhancing the talent acquisition function.

4. Data overload

AI systems are supposed to make hard work easier, not harder. According to a recent survey, the majority of those claiming AI has actually made their jobs increasingly difficult, blame data overload.

“Data for data’s sake is not a solution for smart talent acquisition practices”

Franz Gilbert, Korn Ferry vice president

‍

It’s no secret that AI systems can both produce and analyze huge amounts of data. But finding the implications of that data is what matters.

Supplying recruiters and hiring managers with transparent advice on which candidates to take a closer look at through a user-friendly interface is key for adoption and continued use.

What sectors are using AI recruitment now?

In the past 2–3 years, many have seen the light and jumped on the train towards better recruitment. Here are a couple of examples and what they have to tell about their experience.

Customer support

Amendo has grown rapidly to become one of the most well-known Customer Support recruiting companies on the Swedish scene. With a distinct focus on CS-type roles, they are serving numerous customers from the largest banks to scaling startups. Every year they are filling well over 600 positions. Not only trusted by their many customers, Amendo is a sought-after employer. Recognized by job seekers as a career enabling company that takes care its consultants Amendo attracts a steady stream of job applicants. Working with AI-driven screening has shown very promising effects. The goal for Amendo using AI-screening was to increase efficiency while strengthening the candidate experience and tosupport DEI efforts.

‍

"In our competitive industry we are a well-known brand. Our focus has always been to be the personal brand in which both our customers and job seekers can trust. Not being able to maintain the high quality due to growth and popularity hurts. Hubert helps us stay on track supporting a growing business without sacrificing our employer brand."

Erica Gjälby — Head of Business Management and Customer Success at Amendo

‍

Logistics

Storesupport is one of Sweden's leading brands when it comes to warehouse recruiting. Thanks to their long and successful employer branding campaign, the average job ad now attracts attention from hundreds of applicants.
A dream come true in many aspects, but also a logistical nightmare when it comes to keeping up with the massive influx of applications.

”We save tremendous amounts of time screening candidates as we can focus our full attention on the candidates that we know meet, and are a good match to our criteria.”

Magnus de Woul  —  CEO, Storesupport Warehouse & Logistics

Delivery and transportation

Foodora is Swedens largest food delivery service by far with over 4500 connected restaurants. In entry-level positions where experience isn’t a decisive factor, the volumes easily reaches 200 applications per job or more.

Using an AI-driven recruiting process, Foodora has tackled long waiting times and overworked recruiters while maintaining a great user experience.

“For us, it’s often more important to find candidates who are motivated and driven rather than ticking every box on our wishlist. Evaluating these soft values have traditionally been hard without calling the candidates. With Hubert, candidates are being rated based on soft and hard skills. That helps us find and reach out true stars much earlier than before.”

Olivia Winkvist, foodora Recruiter

Beauty and cosmetics

With about a million applications per 15,000 open positions, L’Oreal has turned their attention to AI as a means of streamlining their recruitment. And it has paid off big-time.

“We really wanted to save time and focus more on quality, diversity and candidate experience. And AI solutions were — for us — the best way to go faster on these challenges”

Eva Azoulay —  global vice-president HR, L’Oreal

‍

Aerospace sector

Airbus was getting questions related to recruiting all 24 hours of the day and had no way of keeping up with requests. The answer was an AI chatbot trained on common questions from historical data which led to a 74% success rate in responding to questions.

“One of the biggest changes we’ve seen is that they [the recruiters] don’t have to answer the same question”

Dave Mills Recruitment Innovation & Airbus Chatbot Product Leader

‍

What do leaders have to say about AI recruiting?

“We have been able to recruit profiles that we probably wouldn’t have hired just on their CV. Like a tech profile for marketing, or a finance profile for sales”

Eva Azoulay, global vice-president of HR, L’Oreal

“Candidate experience and talent identification is a component of our strategy going forward. Chatbots can assist our business source, identify and match talent quickly. We are looking to further leverage chatbots to improve experiences of potential candidates and enable them to meet a consultant as quickly as possible. This will be more profound with our large events when we are required to source several thousand potential candidates”

Giovanni Ambrosini — National IT solutions manager, Adecco

“All of our applicants get a couple of pages of feedback, how they did in the game, how they did in the video interviews, what characteristics they have that fit, and if they don’t fit, the reason why they didn’t, and what we think they should do to be successful in a future application. It’s an example of artificial intelligence allowing us to be more human.”

Leena Nair —   Chief of HR, Unilever

“What took up to a week is now getting done in several hours. The hiring teams think the convenience is simply brilliant. We’ve decreased out time-to-hire from 23 days down to 11”

Ali Ross Grant — Senior Resourcing Manager, Vodafone

‍

How will AI change recruiting practices?

Apart from the obvious that more time can be spent on qualified tasks, there are a couple of less apparent ways in which AI will impact the recruiting business long term.

Increased productivity

A majority of users claim they have increased productivity by between 25–60% just by narrowing the focus of recruiters.

Setting clearer requirements

Getting good results out of AI-based software requires a crystal clear direction from users. More time spent on setting the requirements will exponentially increase the results.

Future recruiters should prepare for a lot more research on what qualities, skills, characteristics, and personalities every specific position requires as so much of the end result depends on it.

Increased focus on team composition

Team research shows that a large part of a group's output depends on the actual composition. As AI-systems can aid in the process of mapping out the individual strengths and weaknesses, there’s room for humans to research deep into team configuration for different functions. Forward-thinking managers are rapidly moving towards this direction and making it their competitive advantage.

New roles to fill: Bias and fraud aversion officers

As more companies place higher trust in AI models, the need to ensure their validity grows stronger. Avoiding bugs and weak points in the hiring algorithms will play a huge role in the success of companies in the future. Ensuring that the traces of bias from training data reduces the effects in the final hire also needs the supervision of competent professionals.

As Manpower SVP Kate Donovan puts it:

"Competitive advantage doesn’t come from technology alone. It also comes from the people who manage it."

‍

Will increased AI adoption mean many recruiters become unemployed?

AI automation may sound like something out of a futuristic movie, and sure we are not there yet. But, take notice; AI will sweep through the world’s economy like a tide wave with big changes as a result. And right now is the best time to start preparing for that shift.

According to SHRM, over 90% of current recruiter tasks will be automated by AI in the coming years. But as experience shows, new tasks emerge from increased technology adoption that will need to be filled. As long as recruiters are willing to adapt to the new landscape, update necessary skillsets, and take on new challenges there will be open positions.

But, this is the future we’re talking about. Currently, only those with low-level repetitive jobs face an immediate threat from tech. The highest efficiency, without question, is still achieved through the beautiful symbiosis between humans and machines working collaboratively.

How will Corona impact adoption of new tech?

The current crisis is horrible for everyone. People are dying, economies are shutting down and employees are being laid off their jobs. But crisis is also the mother of all innovation.

Digital tools have been a big help to keep work going in this unprecedented, global crisis. And in many ways there’s reason to believe that this will speed up technology adoption. As people are forced into using tech in their daily work-life, we believe this will be an eye-opening event with a profound impact on work-life post Corona.

Remote working is one thing that likely will continue to see a much higher use even after things cool down. People realize that remote meetings work pretty well. Twitter recently announced that they will keep allowing people to work from home as much as they like after Corona. As did Facebook. And Coinbase, Shopify, Otis, Square, and many others.

Tobi Lutke, the CEO of Shopify sums it up pretty well in a recent statement:

“Office centricity is over”

Good things often come from trying to do things differently and getting new perspectives. Use of automation for example. Either by pure necessity when parts of the workforce have been let go, or simply to save time.

As the economy picks up again, many teams, especially in recruiting, will find themselves in a position where they are expected to achieve more than before, with even fewer resources to do so. A golden opportunity for automation to step up to the challenge.

Conclusion

AI recruiting tools have a huge potential to transform a major part of the recruitment sector by automating tasks that are too complex for traditional automation. Over time, these systems will likely save valuable resources while boosting efficiency and results.

AI-driven tools are often built on top of existing automation and extend the capabilities in terms of solving the same problems but in a far more refined way. Contrary to what many believe, the candidate experience has lots to gain from AI automation. No, an automated interview will not come close to a human-to-human interview, but when applications are in the hundreds for a single position, the alternative is no interview at all for the vast majority of people.

The potential for reducing bias is also very much plausible, but a word of advice is to be cautious towards vendors claiming that bias is completely removed. The truth can be a long way off, and very complicated to sort out.

As more companies are realizing the potential and adoption soars, regulatory agencies follow close by and are working on policies that will cap the degree of dependency on said systems. If you are shopping for an AI-assisted system, make sure your vendor of choice is compliment with policies in effect, and preferably also those still under review.

When introducing AI recruitment tools in your company, make sure you address aspects such as bias management, deployment time, knowledge, and have a plan for how to use data. These factors have proven to be the biggest hurdles in adoption.

Whether you choose to deploy a system now or later, there’s little doubt that sometime in the near future, you will have to become friends with AI-systems. Technology will keep marching forward, profitability will prevail and you will have to make a choice between adopting new systems or getting left behind in the dust. And Corona will most likely boost the rate of adoption.

When you do decide to take the plunge, know that you are far from alone. A wide array of companies from contact centers to banks and retail are using AI-driven technology to enhance the recruitment process.

Yes, AI will change the way we work, but it will surely be towards the better. With more focus on creativity and social skills and less repetitive work.

When you feel ready, talk with us about setting up a demo and we’ll guide you through all the ins and outs of AI recruitment.

‍

‍

‍