Embracing AI in HR technologies

Artificial Intelligence (AI) is undeniably transforming workplaces globally, and at Simmons & Simmons, we're at the forefront of this revolution with our own Generative AI tool, Percy. We're excited about the transformative potential of AI, but it's important to acknowledge that AI is not merely a future prospect. Many businesses are already integrating AI tools and systems into their daily operations, particularly within the HR function. This article aims to guide senior HR managers, in-house employment lawyers, and other HR technology purchasers through the evolving landscape of AI in HR.

What HR technologies is AI being used in?

Most large employers typically use comprehensive HR systems like Workday Human Capital Management, BambooHR, or SAP Success Factors for many aspects of workforce management from recruitment to payroll and absence management. These systems often serve as the primary repository for employee data and are progressively integrating AI. For instance, large language models are being used to match employee skills to internal vacancies, analyse absence data, and provide chatbots to assist employees. If your organisation uses HR management software, it's likely already utilising some form of AI in the employment context.

The HR technology market is also witnessing a surge in tools focused on specific activities. Recruitment is one area where AI tools have significantly evolved, with systems developed for tasks ranging from CV screening and interview scheduling to background checks and conducting video interviews. AI is also gaining traction in performance management, particularly in roles where performance data is readily available, such as customer service roles in call centres.

Expect to see an increase in HR uses of AI, including chatbots for first-line HR and employee relations support, individually tailored learning and development platforms, annual compensation reviews, and initiatives for equity, diversity, and inclusion.

The changing regulatory landscape: Special protections for employment and HR systems

Lawmakers across the world are proposing and implementing new laws and regulation in relation to the use of AI. Many have indicated that they believe use of AI systems in the employment context warrants special protections. Including:

• The EU achieved a significant milestone last week (8 December 2023) by reaching a political deal on the draft EU AI Act. The Act classifies AI systems in the category "employment, worker management and access to self employment" as high risk AI systems which will need to be assessed before going on the market and throughout their lifecycle (although we note that many such systems are already on the market and in use). Developers of such technologies will be subject to stringent risk management, data governance, transparency and human oversight obligations. In some circumstances, deployers or users of AI systems (such as employers) may be considered the developer under the Act and take on these obligations. The current draft also bans certain uses that may be deployed in the workplace such as emotion recognition systems. The final text of the Act is not finalised or agreed -- but is expected in the first half of 2024. We'll be publishing a more detailed article on implications of the EU AI Act in the employment context in the coming months. Employers should take note that the Act will have extraterritorial application in certain circumstances.

• In the US President Jo Biden issued an Executive Order in October of this year on Safe, Secure and Trustworthy Artificial Intelligence. It included actions related to supporting workers, including specifically to *"develop principles and best practices to mitigate the harms and maximise the benefits of AI for workers"*. In the UK, the House of Commons Science Innovation and Technology Committee called out in its August report the specific challenges for the use of AI in the employment context.

• In China, the State Council issued the Development Plan on the New Generation of Artificial Intelligence involving AI workforce training as far back as July 8, 2017. The plan not only encourages companies and various organisations to provide AI skills training for their employees, but also emphasises the protection of employees who are affected by the use of AI.

In this context, employers should be cautious about reliance on HR systems that may become incompatible with evolving laws (although we expect major vendors will be focussed on achieving compliance where they can keep up).

Existing legal considerations

Whilst we await final regulation, employers should be mindful of existing law that may apply to AI systems in HR. Foremost amongst these are data protection laws. In the EU and UK for example, the General Data Protection Regulation (GDPR) prohibits the making of a decision based solely on automated processing, including profiling, which produces legal effects concerning an individual. The guidelines produced by the UK's Information Commissioner's Office use various employment situations as examples of where this might apply, including recruitment decisions, pay decisions based on productivity monitoring and disciplinary sanctions issued automatically through an attendance monitoring system. Employers should also ensure that they continue to meet the requirements for notifying employees about how their personal data will be processed when new systems are introduced or existing systems are upgraded with new features. Similar rules have been proposed elsewhere, including at state level in the US.

Another key legal concern in most jurisdictions will be the risk of AI systems producing discriminatory outcomes or decisions. Much has been said about the risk of bias in data sets used by HR and employers should seek to understand the data on which systems have been trained. As AI technology develops and data set biases are better understood, AI systems may help employers eliminate human bias and reduce discrimination claims.

What action should we take now?

With the fast-moving legal and regulatory landscape, it can be difficult to understand what action to take right now. The following steps should help ensure current compliance and future readiness for incoming regulation.

1. Inventory: Your organisation is likely already using AI in the HR context. Capture all of your AI use cases and models in an inventory. This is likely to involve speaking to existing vendors. Consider a legal and regulatory risk assessment of these.

2. AI governance framework, policies and protocols: Ensure that everyone in your organisation who might be using or considering HR AI technology understands what they are permitted to implement and when they must get legal, compliance and risk input. For example, would your talent acquisition team seek advice before procuring CV review software?

3. Keep updated on AI regulatory development: Simmons & Simmons' AI View is a fortnightly update on the key legal and regulatory developments across the world. Please get in touch with the authors or your regular contact if you would like to receive this. Follow the authors on LinkedIn for employment specific updates.

4. AI procurement, due diligence and standard terms: Ensure that you are speaking to new vendors of HR technologies about their current and future use of AI. Consider developing standard terms to reflect your organisation's use of AI, to allocate responsibility for ongoing compliance with AI regulation and to ensure liability is properly apportioned.

To help our clients with these and other AI actions, we have developed an AI Toolkit that contains various tools that legal teams can use to deal with AI legal and regulatory risk. Please contact us if you would like to receive the Toolkit. 

Simmons & Simmons AI law practice comprises over 75 lawyers and non-lawyers (including data scientists and AI engineers) across all of its jurisdictions and practice areas, including Employment. We have AI law expertise and experience across our offices in Europe, the Middle East and Asia.