Artificial Intelligence (AI) is now used within the modern workplace for a variety of different functions. From streamlining operations and recruitment, to enhancing decision making and performance management, AI is redefining how employees work and therefore how employers operate.
The introduction of AI in the workplace has raised societal concerns, for example, for vulnerable groups of workers who may find their roles reduced or replaced, but it also raises many other legal concerns (relating to bias, security, use of personal data, copyright and transparency to name a few).
Focusing on the risks associated with employees’ use of AI, these can be minimised by implementing appropriate policies. Such policies can stipulate exactly which AI technology can be used (and therefore which cannot), and how, when and for what purposes they may be used. Employers should also clearly stipulate what data can be inputted into AI, to limit the potential for a personal data breach and disclosure of commercially sensitive and confidential information.
However, one of the most significant risks to employers is the potential breach of employment rights. Most employment legislation requires employers to be transparent in decision making relating to employees and potential employees; if decisions are made by AI, there is a risk that an employer will not be able to explain or justify its decision making. Likewise, many workplace disputes require a reasonable process; this approach will often require emotional intelligence to resolve intricate issues which AI may not (yet!) be capable of replicating.
For example, in recruitment, AI tools can screen CVs to identify suitable candidates and can provide summarised assessments to assist employers determine whether those candidates meet certain criteria. In theory, this should minimise risks of discrimination by removing unconscious bias from decision making. However, there have been various high profile examples of this not being the case. For example, in 2018 Amazon had to scrap a recruitment application after it found that historic data influenced its algorithm, and this resulted in women unintentionally being removed from the candidate pool (this occurred because the majority of people hired for the role in the past had been men so the data the AI trained on led to a conclusion female applicants didn’t meet the selection criteria). Not only would this result in a lack of transparency in decision making, but the results of the AI selection could also amount to direct discrimination on the grounds of sex.
Similarly, facial recognition technology can be used by candidates and employees to access employers’ systems. In theory, this should streamline processes. However, Uber Eats found that the use of facial recognition technology could potentially stop minority ethnic employees accessing their portal, and this could amount to indirect discrimination on the grounds of race.
To harness the benefits of AI while addressing its risks, employers must adopt a balanced and transparent approach. This includes providing appropriate, carefully selected and vetted AI models for use, involving employees in AI adoption strategies, investing in continuous training, and implementing robust frameworks to monitor and review decisions made.
Disclaimer
This information is intended for general informational purposes only and does not constitute legal advice. We recommend seeking professional advice before taking any action on the information provided. If you would like to discuss your specific circumstances, please feel free to contact us on 0800 2800 421.
