At a time when skilled workers are becoming increasingly difficult to find, positions have to be filled more and more quickly and the recruitment process to find suitable employees is becoming more and more complex, the call for support is becoming louder. Be it in the form of suitable technical systems, more personnel or artificial intelligence (AI) or machine learning (ML).
The extended use of technical systems is highly dependent on the company's internal structure and IT and requires precise checks as to whether the systems are suitable for the company. Furthermore, employees who are to work with these systems must be trained and need support, at least in the initial phase. Depending on the size of the company, this can take a lot of time.
There is also the possibility of hiring more staff for recruiting. However, this often fails due to budget and the conviction of the management that there is an increased need for personnel.
When it comes to artificial intelligence, companies are increasingly considering whether it would be a good alternative or complement to the first two options. After all, talking to machines is something many people are now used to from everyday life. We tell Siri or Alexa to play our favourite music or call our girlfriend, and at the mail-order company Otto, chatbot Clara answers all our questions. So why not also conduct job interviews with a robot?
Artificial intelligence (AI) deals with the automation of intelligent behaviour and machine learning. AI is modelled on the human brain and is constantly evolving. It therefore learns, which is why we also speak of "machine learning", or ML for short. The algorithms learn from the existing and ever new data and use it to create analyses that recommend a course of action for the further use of the information. In the next step, AI can also implement these self-generated recommendations for action itself and thus relieve humans of complete areas of responsibility.
AI is already being used in many areas of our everyday lives and makes it possible, for example, to save costs and time or to streamline processes in medicine, in the automotive industry, in agriculture and the energy sector as well as in production.
So it makes sense to use AI or ML to support other areas as well.
The possibilities of AI in recruiting extend across various areas and include, among others:
Some companies already rely on their artificial employees in recruiting. Companies such as Pepsi and Ikea, for example, use "Vera" to select applicants and to conduct interviews independently before the really relevant applicants are selected by "Vera" and handed over to a flesh-and-blood HR employee. The artificial intelligence searches for suitable applicants in job portals, independently conducts video chat or telephone interviews and answers applicants' questions. It gets better and better as "Vera's" algorithm learns more and more with each interview by means of machine learning.
"Vera" is software developed by the Russian start-up Strafory and is currently used in companies that receive several thousand job applications. In these companies, the AI is mainly used for the pre-selection of really relevant applicants, as this initial selection in particular takes a lot of work and time.
Another example is "Matilda". This robot was developed at the University of Melbourne and equipped with 76 questions to conduct 25-minute job interviews after reading application documents. Using AI and ML, "Matilda" is able to read the emotions in applicants' faces and react empathetically.
Whereas "Vera" is primarily intended for jobs such as clerk, waiter or construction worker, "Matilda's" question catalogue is particularly suitable for sales positions. Either way, both AIs must develop further by means of ML in order to actually provide relief in the recruiting departments of companies in the future and to be able to be used for the selection of a wide variety of skilled workers.
At present, AIs such as "Vera" and "Matilda", which perform robotic recruiting, are not yet widespread in German HR departments. In Asia and America, this trend is already much more advanced and has been able to establish itself there in the initial contact with interested candidates. In Germany, chatbots and tools for analysing and matching applications are currently the main tools used. In this way, they offer support to recruiters without completely neglecting the personal aspects of job placements and the candidate experience.
In Germany, Machine Learning (ML) is currently being used intensively in the field of job advertising. Writing and publishing job advertisements is a very time-consuming process within the recruitment process. Especially when working with duplicates, the need for adaptation per job is high - but so is the learning potential of the technology. This makes it an ideal field of application for artificial intelligence, because in these recurring processes, the ML algorithms can learn very quickly and also perform comprehensive, predefined processes such as "gender bias search" automatically.
The sourcing of CV databases can also be automated by using the learning potential of machine learning. Through the automation and concrete description of the requirement profile, the most suitable female candidates can be found in this way. Compared to non-automated search by recruiters, this is a time-saving and effective solution.
The advantages of AI in recruiting are not only the many thousands of conversations an AI can have per day and the countless job ads it creates and optimises. It also doesn't get tired or sick, speaks different languages and can work around the clock regardless of time zones. A recruiting robot can also adjust gender and thus fully adapt to the requirements of the candidates. Furthermore, robots do not discriminate and do not make decisions "on instinct". In particular, "unconscious biases" - unconscious, prejudiced tendencies - which every human being carries within him or her, can significantly reduce discrimination in recruiting processes through the use of robots, as they analyse objectively and neutrally and are not aware of any prejudices or antipathies.
As already described, AI is characterised above all by constant learning. It therefore learns more and more about the candidates and is able to find out more information about applicants on the internet within seconds and put it together. As a result, the AI has a more comprehensive picture of an applicant than flesh-and-blood personnel. This means that AI can possibly judge even better whether a candidate is a good fit for the company or not.
But isn't it precisely the variance of employees that makes companies efficient and creative in the first place? The danger that companies will then only consist of a "one-size-fits-all" workforce in the future is definitely present with the extensive use of AI. How are constructive discussions and conversations supposed to develop if all employees fit together ideally?
Another downside is data protection. Human resources managers are particularly debating whether it is still in line with data protection regulations if a robot voluntarily and involuntarily collects published data about applicants on the internet and uses it to make a job decision. This is another reason why many companies are still critical of the use of AI in recruiting departments.
Every company has to go with the digital transformation in order not to be left behind in the highly competitive labour market. Suitable skilled workers are becoming harder to find, which makes the recruitment process more difficult, more time-consuming and requires creative ideas from HR professionals. Simply advertising a job is no longer sufficient. The relevant professionals must be actively sought and approached. Thus, the decision for support by artificial intelligence is only a matter of time.
AI, with its extensive and objective analysis possibilities and also as a chatbot, should indeed be used as a support, but it will never replace personal contact between recruiters and applicants.
ARTS also uses a matching tool to select applicants and suitable job offers. However, the final decision is still made by the flesh-and-blood HR team. If you find an interesting job on our job board and you would like to discuss it, feel free to contact our team, apply directly for a job or send us an unsolicited application.