The often-expressed fear that AI robots are on the verge of infiltrating and taking control of every aspect of our lives is admittedly understandable, given the capabilities of AI already proclaimed today: writing guest articles for newspapers, answering basic customer service queries, diagnosing medical conditions, solving long-standing scientific problems in biology and much more – if you believe the sources [1],[2]. But are these purported successes really evidence of unlimited potential? Will AI systems really be able to solve any task given enough time and data? You may not want to hear this, but the answer is a resounding NOPE. At least not if researchers and developers stick to the knowledge-lean approaches they are currently so fixated on.

To start with, artificial intelligence is not even remotely comparable to human intelligence. Any AI system is a sophisticated ‘bag of tricks’ designed to lull humans into believing it has some kind of understanding of the task at hand. So, to develop AI-based technologies that are intelligent in any sense of the word there is no way around feeding these technologies with extensive human knowledge – which involves a substantial amount of human labor and will do for the foreseeable future. As a consequence – and again, you may not want to hear this – relying on solutions for HR and labor market management based solely on deep learning (DL) or other statistical/machine-learning (ML) approaches is a bad investment. And it is not simply a waste of money, it is also an ethical issue: Especially in the field of HR, the reliability of data and evaluations is crucial as it can deeply affect human lives. For instance, all too often perfectly suitable candidates are screened out by AI-based systems like ATS just because their resume does not contain the exact keywords specified in the filter or is associated with false contexts. Which a human recruiter would have realized, had they seen the resume themselves. This is just one of many examples of how real people can be affected by underperforming AI technology.

Artificial – as in fake

While researchers define AI systems as ones that perceive their environment and take actions that maximize their chance of achieving their goals, the popular perception of AI is that it aims to approach human cognition. Intelligence is typically defined as the ability to learn, understand, and form judgments or opinions based on reason or to deal with new or trying situations. However, this requires a key cognitive ability: storing and using commonsense knowledge, which we humans develop through a combination of learning and experience – and that so far, AI systems simply do not have and won’t achieve in the foreseeable future. These limitations are most obvious in natural language processing (NLP) and natural language understanding (NLU) techniques based on ML because commonsense knowledge is absolutely essential when it comes to understanding natural language. As an example, consider the following statement:

Charlie drove the bus into a tree.

Nowhere in this sentence does it explicitly state that Charlie is a human being, was in the bus, or that this is uncommon behavior. And yet our commonsense knowledge allows us to draw these and many other conclusions from this simple sentence without much effort. This ability, coined ‘linguistic competence’ by linguist Noam Chomsky, distinguishes computer systems trained in NLP and NLU fundamentally from human cognition. While we humans acquire this linguistic competence at an early age and can use it to discern the meaning of arbitrary linguistic expressions, knowledge-lean AI models will never be able to do so to the same extent because they work on a purely quantitative basis: their ‘intelligence’ is based on statistical approximations and (occasionally pointless) memorization of text-based data. ML systems can, at times, sidestep the problem of understanding and give the impression that they are behaving intelligently – provided they are fed enough data and the task is sufficiently narrowed down. But they will never actually understand the meaning of words; they simply lack the connection between form (language) and content (relation to the real world) [1].

This is precisely why even the most advanced AI models still struggle with these types of statements: because they contain so much implicit, often important information and causalities. For example, GPT-3, a state-of-the-art AI-based language model (which wrote the newspaper article cited at the beginning), was unable to correctly answer the simple question of whether a toaster or a pencil was heavier [1]. This is somewhat reminiscent of a quote from Oscar Wilde’s The Remarkable Rocket: “I am so clever that sometimes I don’t understand a single word of what I am saying”…

A major reason for this problem is that commonsense knowledge comprises an unconceivable number of facts about how the world works. We humans have internalized these facts through lived experience and can use them in expressing and understanding language without ever having to encode this staggering amount of knowledge into a written form. And precisely because this tacit knowledge is not captured systematically, AI systems have no access to it – or at least knowledge-lean AI systems don’t, i.e., systems based purely on statistical/ML approaches. So these systems are faced with unsurmountable challenges when tasked with understanding language. Because it is ‘unexpected’.

Another simple example: In a statistical analysis of words related to the English word pen, an ML system may spit out the words Chirac and Jospin, because these names are often mentioned together with the French politician Marie Le Pen, who of course has nothing to do with writing tools. It gets even more complicated when the same expression takes on different meanings depending on the context – think writing pen versus sheep pen. Systems based purely on ML often have great difficulty in discerning the nuances of such everyday language because they do not store the meanings of a word; connections are just based on cooccurrence. So, in the knowledge-lean world, there is still a long way ahead to reliable NLU.

No AI without HI

Having been around since the 1950s, AI has cycled through phases of hype and disillusionment many times. And right now, at least in the subfield of NLU, we are cycling back into the ‘trough of disillusionment’, as Gartner has so aptly coined it. Nevertheless, many are still clinging on to the great promises, blithely publishing, touting and investing in knowledge-lean AI technologies. But relying completely on ML-based algorithms for any application that requires language understanding is nothing but an expensive mistake. As we already explained, it is a huge leap from automated processing of textual data (NLP) to meaningful human-like understanding (NLU) of this information by machines. Thus, many automation plans will remain an illusion. It is high time to switch to a strategy that can succeed in these challenging tasks by effectively creating artificial intelligence through human intelligence.

In our area of expertise here at JANZZ, where we (re)structure and match job-related data, we understand that many automated tasks in big data require a significant amount of human labor and intelligence. Our job and resume parsing tool JANZZparser! has relied on NLP and NLU since the beginning – but always combined with human input: Our data analysts and ontology curators carefully and continuously train and adapt the language-specific deep learning models. NLP tasks are trained using our in-house, hand-selected corpus of gold standard training data. Parsed information is standardized and contextualized using our hand-curated ontology JANZZon!, which is the most comprehensive multilingual knowledge representation for job-related data worldwide. This machine-readable knowledge base contains millions of concepts such as occupations, skills, specializations, educations and experiences that are manually linked by our domain-specialized experts according to their relations with each other. JANZZon! integrates both data-driven knowledge from real job postings and resumes as well as expert information from international taxonomies such as ESCO or O*Net. This is the only way to ensure that our technologies can develop the kind of language understanding that actually deserves the name artificial intelligence. Generic phrases such as flexibility are given the relevant context, be it in terms of time management, thinking, or other aspects. As a result, false matches such as Research and Ontology Management Specialist with occupations like those in the figure below, due to overlap in wording but not in content, are excluded from matching results in our knowledge-based systems. The unique combination of technology and human intelligence in machine-readable form can achieve highly accurate, reliable and cross-linguistic/cross-cultural results when processing job-related data. Errors like the one in the pen example simply do not occur because each word is conceptually linked to the correct and relevant meanings and points of association.

Throwing good money after bad

The fact that we are on the right track with our hybrid, knowledge-based method of combining human intelligence with state-of-the-art ML/DL methods is not only confirmed by our own experiences and successful cooperation with businesses and public employment services (PES) across the globe, but also widely recognized by – non-commercial – NLU researchers. The outlined problems around the missing cognitive component in knowledge-lean AI systems will not be resolved in the next 50 years. As soon as human language is involved, there will always be countless cases where a 3-year-old child can make the correct semantic connection while a machine-learned tool either fails or does so only with absurdly high effort. Although knowledge-based systems like ours provide reliable and explainable analysis of language, they fell from grace because researchers and developers perceived the manual effort of knowledge engineering as a bottleneck. And the search for other ways to deal with language processing led to the knowledge-lean paradigm. Nowadays, supported by the immense speed and storage capacity of computers, most have come to rely on applying generic machine learning algorithms to ever-growing datasets for very narrow tasks. Since this paradigm shift, many developers and consumers have invested a lot of time and money in these systems. Being so heavily invested financially, they are simply not prepared to admit that this approach cannot produce the results they are looking for, despite the growing evidence against them.

However, the hybrid, knowledge-based approach of combining ML-based features with human-generated semantic representations can significantly improve the performance of systems that depend on language understanding. In our case, by adopting this approach, our technology can avoid the pitfalls of knowledge-lean systems based on uncontrolled AI processes, simple keyword matching and meaningless extractions of intrinsically context-poor and quickly outdated taxonomies. Instead, our matching and analytics solutions can draw on the smart data generated by our ontology. This context-based, constantly updated knowledge representation can be used in a variety of ways for intelligent classification, search, matching and parsing operations, as well as countless other processes in the area of job-related data. Especially in HR analytics, our solutions achieve above-average results that far exceed the performance of comparable offerings on the market. Thanks to these insights, employers are able to make well-informed decisions in talent management and strategic workforce planning based on smart, reliable data.

Do the right thing and do it right

Finally, there are the ethical concerns of applying AI to textual data. There are numerous examples that illustrate what happens when the use of machine learning systems goes awry. In 2016, for example, a software manufacturing giant’s chatbot caused a public controversy because, after an unsolicited, brief training session by Internet trolls, it almost immediately started spouting sexist and racist insults instead of demonstrating the company’s NLP technology in an entertaining and interactive way as planned. The challenge of developing AI that shares and reliably acts in accordance with humanity’s moral values is an extremely complex (and possibly unsolvable) task. However, given the trend toward entrusting machine learning systems with real-world responsibilities, this is an urgent and serious matter. In industries such as law enforcement, credit or HR, the inadequate use of AI and ML is all the more delicate. Talent and labor market management, for instance, directly affects the lives of real people. Therefore, every decision must be justifiable in detail; faulty, biased or any kind of black-box automation with a direct impact on essential decisions in these matters must be weeded out. This stance is also taken by the European Commission in their whitepaper on AI and the associated future regulations, especially in the HR sector. As a matter of fact, almost all of the highly praised AI systems for recruiting and talent management on the market, mainly originating from the USA, would be banned under these planned regulations. JANZZ.technology’s approach is currently the only one that will be compatible with these planned regulatory adjustments. And this has a great deal to do with our knowledge representation and how it allows us to produce not just AI technology that comes very close to understanding language, but in fact explainable AI. So ultimately, the way forward is to appreciate that – in the words of NLU researcher McShane – there is no bottleneck, there is simply work that needs to be done.

At JANZZ.technology, we have done this work for you, with experts from diverse backgrounds in terms of language, experience, education and culture. Their pooled knowledge is incorporated into our ontology JANZZon! and made readable and processable for both machines and humans. Together, our experts have created and continuously curate the best possible and most comprehensive representation of the ever-growing heterogeneity of job-related knowledge in the field of human resources and labor market administration. Enabling multilingual, modular and bias-free solutions for all HR processes – and bringing you a step closer to truly intelligent HR and labor market management solutions. If you would like to learn more about our expertise and our products, or benefit from advice tailored to your organization’s individual situation, please contact us at info@janzz.technology or via contact form, visit our product page for PES and listen in on our new podcast series.

[1] Toews, Rob. 2021. What Artificial Intelligence Still Can’t Do. URL: https://www.forbes.com/sites/robtoews/2021/06/01/what-artificial-intelligence-still-cant-do/amp/
[2] GPT-3 (Guardian). 2020. A robot wrote this entire article. Are you scared yet, human? URL: https://www.theguardian.com/commentisfree/2020/sep/08/robot-wrote-this-article-gpt-3

Does this situation seem familiar to you? On the Uber ride to the airport, you get caught up in a conversation with the driver and before you know it, you’re in the middle of a discussion about the potential of genetically modified bacteria to create cancer drugs. It quickly turns out that there is an extremely educated person in the driver’s seat who is currently putting his education to rather limited use. A new report from the International Labor Organization (ILO) on this matter now shows that only half of all workers worldwide have an occupation that matches their level of education. At the same time, many employers report difficulties in finding qualified personnel. This situation not only points to a significant gap between educational institutions and the working world but, depending on the facts, it also undermines the relevance of omnipresent keywords such as ‘upskilling’, the hype of which we have already questioned in the past.

The Report

The main findings of the report published by the ILO in September reflect realities that we, at JANZZ, have been anticipating for some time. Despite improved access to education and an increased level of education, the overall match between workers’ level of education and the actual skill level required for their occupation is a sobering 50% worldwide. Breaking this mismatch down into its individual components results in over- and under-qualification situations. [1] In concrete terms, this means that oftentimes there is a gap between the work that people can or want to do, and the job profiles and qualifications for which there is an actual demand. Indeed, from our own data analysis, we can cite as an example the fact that, in Europe, there is currently a shortage of about 400,000 truck drivers, as well as hundreds of thousands of nursing and care workers and employees in the catering and hotel industry. The main problem here, however, is not a lack of skills but rather a lack of willingness to learn certain professions. The reality is that the previously mentioned occupational fields are often underpaid, thus, resulting in a general lack of interest to then remain in those positions for a longer period of time.

Accordingly, the ILO study found that while there is over- and under-skilling in all countries, the patterns differ depending on the income level of each country. The matching rate increases with progressively higher median income in a country, which means that in countries with lower wages, just one in four workers have a job that correlates with his or her level of education, like our opening example of the biotechnologist driving a cab. There is a tendency for undereducation to be more common in lower income countries, while overeducation is more common in high income countries. [1]

There are different reasons for both circumstances: Workers are sometimes overqualified for their jobs because they accepted them for motivations other than aspiration, for example because the position offered specific benefits such as shorter commute times, better work-life balance, or an experience opportunity for later, more demanding jobs. However, these people are only part of the workforce. Another part can be attributed to distortions in the labor market, according to which the supply of workers with higher levels of education exceeds their demand. In the case of undereducation, the main causes are in turn the relatively low completion rates and the lack of formal qualifications amongst a large workforce, which is more common in low-wage countries where educational opportunities tend to be poorer. [1] In addition, given the low chances of finding a job in their own level of education, many also take available jobs for which they are overqualified, but which provide them with an immediate source of income for their livelihood. From our own sources, we outline the example of Paraguay, where around six times more lawyers are trained than the labor market can accommodate after their graduation. Many of them end up in jobs in call centers, retail or administration, where there is hardly any connection to their legal skills and knowledge.

Among the inadequately trained employees, some can at best make up for the lack of qualifications by acquiring the necessary skills through on-the-job training or self-study. Overall, however, there are still numerous negative consequences of these mismatches due to over- or under-qualification that affect the workforce, the employer community, and society as a whole. For example, a high degree of overqualification can lead to a loss of motivation and suboptimal returns on investment in education and training. In addition, every overqualified person always occupies a position that could or should be held by someone more suitably qualified. On the other hand, many underqualified employees face difficulties in the transition from the informal to the formal economy, which can have negative consequences on productivity and innovation, as well as economic growth. Since the examples that we mentioned are rather longer-term phenomena, according to the ILO, systematic, i.e., political measures are usually required to solve them. [1]

The problem of upskilling and reskilling

Let us make a few more comments on one specific finding of the ILO study. Among other things, the report emphasizes that in countries with medium-high to high average incomes, about one-fifth of all workers today are overqualified [1]. In countries like Finland, where just under 75% of the population has a university degree but where there are considerably fewer jobs at this level of education, this rate is even higher [2]. At the same time, keywords like ‘upskilling’ and ‘reskilling’ are on the lips of everyone who talks about the development and future of labor markets. Such training processes can make sense within a company, but they are not automatic, especially not when it comes to the entire labor market of a state or region. As soon as experienced employees who are well established in their positions are moved or promoted to another function or activity because of upskilling, their old position must first be filled with someone suitable. This task alone can cause red flags for HR, as the assumption that all such ‘pre-upskilling’ positions will immediately cease to exist is misleading and incorrect. The situation may become even more convoluted when a country has a harsh migration policy and there is a lack of workers for low-wage jobs (often disdained by citizens), for instance, as can be observed nowadays in the nursing field in the UK [3].

Furthermore, in his or her new position, it is not certain that the promoted person will perform equally well or be as satisfied as before, or whether the promotion and the accompanying increased responsibility is a right fit for that person. Management literature knows the so-called Peter Principle, according to which a large part of the inefficiency on the labor market is due to the fact that in hierarchical structures, such as a labor market, every employee tends to rise to the point of incapacity. Although this theory has a satirical undertone, it nevertheless seems to be somewhat true, not least when considering the rather low downshifting rates [4]. When it comes to reskilling, the situation looks somewhat better in that it is conceivable in this process to be able to transfer and adapt at least so-called “core skills” from the old job to the new one.

Back to the findings of the ILO report: Even if none of the above problems occur, the fact remains that in our geographical location many people are overqualified for the work that they do on a daily basis. A prime example of this is the trend among medical students who seek to become a specialist, even though, depending on the area, the occupational demand for it can be rather limited and there is an existing deficit of family doctors. Moreover, one can also wonder, why at Swiss universities, efforts are being made to attract not only geography students but also lateral entrants from geology studies to train as geography teachers for secondary level II. What is problematic about this is that there is already no (increasing) demand for this profession, while at the same time, there is also a shortage of primary and lower secondary level teachers. Furthermore, data from an OECD survey indicates that, compared to EU/OECD countries, Swiss nationals are more often overqualified than immigrants, with the latter accounting for an above-average share of the workforce in the low-wage sector (60%) [5]. This also indirectly reflects where there is a real need for personnel…

Quo vadis?

Over- and under-skilling remain a current problem in most labor markets, despite considerable progress in global access to education. Both conditions reflect underutilization of human capital and can carry high economic and social costs. What needs to change?

To reduce such mismatches overall, it is first necessary to capture and assess the extent to which the level of education of workers in a labor market matches the level of education required for their jobs. This is done by means of analyses such as the ILO’s cited here. (See also our English-language contribution to a Nobel Prize-winning paper that systematically explains the asymmetry between the large number of unfilled jobs and simultaneous unemployment within a market). As a next step, this data would also need to be incorporated into a country’s education planning and human resource development, which may be complemented by policy measures. In countries with high rates of undereducation, the undereducation of those who already occupy highly skilled jobs or will do so in the future must be raised by qualitatively adequate means. It should be noted, however, that simply upskilling all job seekers in a labor market that offers many low-skilled positions will neither automatically eliminate the mismatch, nor lead to a reduction in the unemployment rate.

Furthermore, it would be important to actively create more transparency for the public about the actual demand of a labor market, especially in countries with a high rate of overqualification. Ideally, such information would motivate future workers to train in a ‘meaningful’ direction or at least to be aware of the suboptimal employment opportunities after graduation and to inquire about real (and often equal in terms of wages in German-speaking countries) alternatives, such as apprenticeships. What is therefore certainly not conducive is to propagate upskilling and reskilling towards trendy degrees and competencies as ‘the thing to do’, which unfortunately also happens regularly today due to published (and quoted) misinformation on this topic. The latter can have rather dangerous and far-reaching consequences for society, especially if done through governmental and political voices, because such false promises and forecasts will widen the skills gap more and more and possibly have global implications in areas such as migration.

Finally, more emphasis should be put on adequacy in the actual matching process between candidates and jobs. This requires, first and foremost, reliable and information-dense data collection, analysis and classification. Here at JANZZ, we gather just such information through a variety of projects, including collaborations with the Public Employment Services (PES) of countries around the world. This has allowed us to develop market-leading evidence-based solutions since 2010. Our systems are not only efficient, scalable and extremely powerful, they also rely on ontology-based semantic matching. Furthermore, our tools all deliver unbiased results according to the OECD principles on AI. We are keen to stimulate fact-based discussion on all issues related to labor markets and processes, and to raise social awareness about them. After reading this article you might think about how many people in your environment are working in a profession that truly corresponds to their degree (level)…

If you would like to learn more about our offerings, please contact us at info@janzz.technology or via contact form, or visit our product page for PES.

[1] ILO. 2021. Only half of workers worldwide hold jobs corresponding to their level of education. URL: https://ilostat.ilo.org/only-half-of-workers-worldwide-hold-jobs-corresponding-to-their-level-of-education/

[2] Clausnitzer, J. 2021. Population with educational qualification in Finland 2019, by level of education. https://www.statista.com/statistics/528083/finland-population-with-educational-qualification-by-education-level/

[3] Inman, Phillip. 2021. Does the UK have a wage problem? URL: https://www.theguardian.com/money/2021/oct/06/uk-wage-boris-johnson-skilled-skilled-economy

[4] Donzé, René. 2021. Mehr Leben, weniger Hamsterrad: Wieso die wenigsten einen beruflichen Neustart wagen. URL: https://nzzas.nzz.ch/spezial/downshifting-wieso-nur-wenige-einen-beruflichen-neustart-wagen-ld.1612299

[5] Loos, Melanie. 2018. Schweizer öfter überqualifiziert als Zuwanderer. URL: https://www.handelszeitung.ch/konjunktur/schweizer-ofter-uberqualifiziert-als-zuwanderer