Artificial Intelligence (AI) and employment law: 6 key considerations

AI presents enormous opportunities for businesses who embrace it, but also very significant challenges – including in relation to how employees are engaged and managed. In this article, we explore six key areas and consider the impact of AI on employment law and HR. Is it time to update your ‘to do’ list?

AI will change the way that many of us work – and there are huge opportunities for businesses who embrace AI to make better decisions more quickly. However, it is not without its challenges and there are no laws specifically for the purpose of governing the use of AI at work. The impact of this is that regulation of this area is governed by existing employment legislation, much of which predates the iPhone, and which was not designed to address the rapid development of AI.

This means that employers must continually update their knowledge of the opportunities and challenges that AI brings and consider these in the context of existing employment laws and their own ethical principles where there are gaps in the legislation.

AI presents enormous opportunities for businesses who embrace it, but also very significant challenges – including in relation to how employees are engaged and managed. In this article, we explore six key areas and consider the impact of AI on employment law and HR. Is it time to update your ‘to do’ list?

This presents a significant challenge for employers and HR leaders. As a starting point, we have identified six key areas for employers’ HR leadership to consider in the context of AI and their workforce:

1. Could AI increase the risk of discrimination and bias?
2. What’s the impact on transparency in employee communications?
3. How does AI interact with data protection and privacy?
4. What about job displacement and skills development?
5. Will AI impact how we define and engage employees?
6. How can we demonstrate ethical use of AI?

AI is already being used by some businesses to support decisions on hiring. We anticipate that this trend will continue, with AI being used to support performance reviews and other HR functions. This has the potential to free up huge amounts of management time and, in theory, allow businesses to make HR decisions based on objective data and criteria, with fairer results.

The potential issue is that AI program have been found to exhibit or even learn bias, in some cases despite safeguards being put in place to minimise this risk.

While programmers may get better at avoiding this, ultimately if an individual is not selected for recruitment or receives a low performance grade as a result of a biased AI output, it is the employer who will be liable for a discrimination claim. Even more damaging could be the reputational damage and loss of trust from employees.

The solution to this challenge will be to couple any use of AI in management decisions with robust oversight and challenge. This could be via robust appeal mechanisms overseen by employees and ongoing testing of the output of the AI algorithm to try to identify and remove any indications of bias.

Another consideration when using AI in respect of HR decisions is maintaining your workforce’s trust. This will be undermined if employees are not taken along on the journey and they do not feel that they understand how the AI is being used.

If AI is used, it will be important to be able to explain to employees how it has been integrated into decision making processes and how the business is able to be accountable for the decisions driven by AI.

Employers who embrace AI will need to introduce specific AI policies setting out the detail of how they use AI and what safeguards they have put in place regarding its use.

Related to transparency, employers will need to consider the data protection implications of using AI, including ensuring that the employee data which feeds the AI has been properly obtained and that the use of the data in this way is permissible. Failure to do so could lead to significant fines and reputational damage.

Employers planning on using AI should review their data protection policy and privacy notices to ensure that they are legitimately processing employee or candidate data.

Where the AI solution is provided by a third party, employers will also need to be alert to their data being used by the third party for providing services to other employers. Reviewing and understanding commercial terms will be key.

The first three points dealt primarily with the implications of using AI in HR applications. However, one of the most controversial impacts of AI for employees will be the displacement of jobs or skills.

The potential for job automation in certain industries will mean that fewer employees will be required to carry out certain tasks. To be prepared for this, employers need to consider upskilling or reskilling employees to adapt to AI and tech advancements and have a plan for this.

This will also need to be carefully balanced against the need for an employee pipeline. Where more senior roles still need to be carried out by employees, if the routes to obtaining those roles through junior level jobs disappear, businesses will need to develop new ways of training the next generation of employees to run the business.

Much of the case law regarding the definition of an employee is from the 1800s, with the concept of the ‘master and servant’ still prominent. As we start to see more tasks carried out by AI, supervised by human employees, it is also possible that the way we define the employment relationship could change, with a potential move away from the prevalence of ‘personal service’ as a determining factor for employment status.

In addition, we have seen a trend towards more flexible and informal employment as a result of the expanding ‘gig economy’. This could be accelerated as more workers are engaged on an ad hoc basis (perhaps working for multiple businesses) to manage AI, rather than having a dedicated job for one employer.

As set out at the start of this article, there is currently very little regulation around the use of AI and so, in the absence of regulation, employers will need to develop their own principles for how and when they will use AI.

Key principles to consider:

1. Is this an area where I should be using AI? 

There are huge potential benefits to AI, but businesses will need to make a value-based decision on not only can it be used for a particular task, but also is it fair on my workforce to use it for that task. Particularly controversial areas (especially given the risk of bias) are the use of AI in disciplinaries, grievances and performance management. This is also not a question that can be answered at a single point in time and will need to be continually reassessed as the technology and attitudes towards it change.

1. Does the way I am using AI protect the dignity of employees and maintain the organisation’s ethical principles? 

At its simplest level this might simply mean ensuring that the use of AI is compliant with existing policies and procedures (or amending those procedures if required). However, employers should also think about their broader corporate values and how the use of AI aligns with them.