Artificial Intelligence (AI) has already started to revolutionise workplaces and is affecting both employers and employees in many ways. The purpose of this blog post is to provide an overview of some of the workplace areas which are significantly impacted by AI, from recruitment through performance assessment to remote working and robotics, and to set out the benefits as well as some potential dangers connected with each of them.
AI in Hiring and Recruitment
AI-based application tools are widely used to improve the hiring and recruitment process. Algorithms can help companies to write more compelling job descriptions and to identify suitable candidates for a job opening in an efficient and cost-effective way. Textio, for example, is an augmented writing tool that analyses job descriptions and suggests alternative wording that is neutral in tone in order to ensure gender balance and target the most qualified candidates. AI-based application tools can also help companies to eliminate bias and discrimination in the hiring process by making objective recruitment decisions based on skills rather than on demographic or personal bias. By programming an algorithm with legally acceptable questions only, the risk of asking prohibited questions in a pre-hire interview, such as questions regarding religious beliefs or sexual orientation, can be reduced.
However, the use of AI-based application tools in the hiring process also brings with it certain dangers. There is a risk that algorithms do not eliminate discrimination in the hiring process, but instead perpetuate existing biases or form new biases based on the initial dataset provided. If the data provided to the algorithm is unfiltered, uncleansed or unrepresentative, these limitations will be built into the AI-system. If, for example, the training data to evaluate candidates to fill management positions within a large organisation is based on successful male candidates from prior years, the output will likely be biased against women. There is also a risk that algorithms are programed by people who are biased themselves and as a result that bias could be inbuilt into the foundations of the recruiting AI-system.
Another risk of AI-based application tools is the breach of privacy rights of job candidates, in particular where algorithms are used to assess the emotional intelligence of candidates, such as by measuring eye movements, facial gestures or even heart rates. Under Swiss law, an employer may only process data of employees to the extent that such data concerns the employee's suitability for his/her job or is necessary for the performance of the employment contract. Further restrictions apply in relation to the processing of data concerning the health of employees (such as the measuring of heart rates) and other sensitive data. Data must be used only for the purposes apparent to the applicant at the time the data is collected and the data must be deleted when no longer needed.
AI in Performance Assessments and Development of Employees
Employers can use AI to monitor and assess the performance of employees. The resulting data can assist employers in a variety of ways, including in employee performance reviews. Decisions about bonuses, benefits, salary increases or promotions, for example, could be made based on actual employee output which is tracked by technology. Such performance reviews using AI can prevent discrimination in the workplace and increase equality by removing unconscious bias and relying on factual information only. Furthermore, by monitoring employee performance, AI-based tools can learn how different employees carry out specific tasks and subsequently assist in the onboarding of new employees and in the training of existing employees in how to do things more efficiently.
However, the monitoring and tracking of employees in the workplace can increase occupational health and safety risks. Increased monitoring can lead to micromanagement of employees, which is a prime cause of stress and anxiety. For example, employees may feel pressured to improve their performance and as a consequence begin to overwork.
Another danger of monitoring and performance tracking in the workplace is the breach of privacy rights of employees. Swiss labour law restricts monitoring and control systems which are designed to monitor the behaviour of employees in the workplace and requires that every monitoring measure is justified by an overriding interest of the employer, such as occupational safety, quality assurance or the optimisation of work organisation. The measure must be designed and implemented in a way that the health and freedom of movement of employees is not affected and there must be no less intrusive measure available to achieve the same result.
Robotics and Automation in the Workplace
Particularly in the industrial sector, the introduction of robots and automation in the workplace has led to an optimisation of work and production processes, a reduction of costs and an improved overall productivity of companies. In addition, companies can improve occupational health and safety by using robots for dangerous or physically stressful tasks. Another benefit of robots and automation is the improvement of work satisfaction of employees. By releasing employees from repetitive or mundane tasks, employees can focus their energy on areas which are better suited for them, such as on more creative tasks or tasks that require human interaction.
However, also here the use of robots in the workplace brings with it certain dangers. There is an increased robot-human collision risk, in particular where robots overcome safety barriers and collaborate with humans (so called collaborative robots). Accidents can result, for example, from system failures, a reduced alertness of employees or due to risk compensation. This also creates a new challenge for the employer who under Swiss law has a duty to safeguard the personal safety and health of the employees and has to take all measures that are shown by experience to be necessary, feasible using the latest technology and appropriate to the particular circumstances of the workplace to prevent occupational accidents.
AI-based Tools in the Workplace
AI-based tools have innovated and facilitated the work of employees across industries and professions. In the healthcare sector, for example, AI helps to detect diseases more accurately and in the early stages and assists doctors in the prescription of the best treatment. AI has further assisted banks and other financial institutions in their financial crime and KYC (Know your client) compliance. Encompass, for example, is a KYC software which gathers and analyses a range of information and provides a full picture of a new customer within minutes. With such tools, customer onboarding can be completed in a timely manner, with a high standard of due diligence and at a decreased cost.
As well as the challenge of changing mentalities to accept the enhanced insights that such systems can provide and of putting in place the required level of information and consent, particularly where personal data is involved, adopting AI-based tools in sensitive environments such as hospitals or banks requires careful allocation of responsibilities and of liability if something goes wrong. For regulated activities there are even more hurdles, as the regulator will generally insist on transparency of decision-making, which an AI-tool cannot give.
AI for Remote and Flexible Working
AI has further allowed companies to create virtual workplaces that can be accessed by employees from anywhere with an internet connection. AI-tools such as Google Hangout create collaborative workspaces that allow for group messaging, video calling and desktop sharing giving employees outside the workplace the opportunity to engage with and contribute to work in a meaningful way. This is particularly beneficial for certain groups of employees such as parents, who have to balance other responsibilities at home, and disabled people. Virtual workplaces can therefore lead to less discrimination and more gender equality in the workplace. For the employer the benefits of remote and flexible working include reduced expenditures, increased productivity, efficiency and performance from the employees with limited disruption from external factors, such as weather conditions. In addition, companies can extend their business hours.
However, there is a danger that working remotely does not increase employee efficiency but rather reduces employee engagement due to isolation and lack of structure. There is also a danger that employees fail to disconnect from work even when their working day is done, which can damage the employee's health in the long run. As employers have a duty to ensure the wellbeing of their staff, these risks to mental health must be mitigated by the employer to the extent reasonably possible. The employer's legal obligations to ensure working time is recorded and that the working day is not too long and with proper breaks should also not be forgotten.
AI in the workplace is here to stay. To gain the most from the benefits that it can bring, it is important to also be aware of the risks and to take measures to remove or reduce them wherever possible.
Feel free to download this blog post as a pdf file by clicking the button below.
To learn more about the legal aspects of AI and what to look out for when implementing AI solutions in your company sign up to our Blog.
Photocredit: Drew Graham, Unsplash