How AI is changing global payroll and HR management

Artificial intelligence has taken center stage in the business world ever since the surge of generative AI. Chatbots, digital assistants, and other types of smart technologies are already being used across industries and business functions, and payroll and HR are no exceptions.

Gartner reports that 76% of HR leaders believe that their organization will fall behind if they don’t implement AI solutions in the next year or two. This fear of lagging in organizational success translates into a rush to explore potential use cases. In fact, 34% of HR leaders were already looking at options for using generative AI when questioned in a Gartner benchmarking session earlier this year.

As an international keynote speaker, author, and HR & Payroll Tech Expert, Anita Lettink advises multi-country organizations, vendors, and investors on selecting, building and investing in payroll and HR solutions and closely follows the adoption and implementation of AI in HR and payroll.

We’ve had a chat with her about how AI is changing global payroll and HR management. Read on as we dive deep into the prerequisites for using AI, the challenges businesses need to overcome, the critical role tech vendors play in keeping their clients up to date, and much more.

Lano: Hi Anita, great to have you. Let’s start with a generic question. In your opinion, what are the current main struggles businesses face in global HR and payroll management and how can AI help them overcome these struggles?

Anita Lettink: I'm going to take one step back because not every organization has HR and payroll processes that are organized at a global level or has migrated to the cloud for HR and payroll. When clients approach me, they often have a scattered payroll and HR landscape with all kinds of different solutions, and they are in different stages of maturity. This kind of set-up makes it very hard to do anything with AI because there are several conditions that need to be fulfilled so that organizations can use AI in a way that really benefits them.

"There are several conditions that need to be fulfilled so that organizations can use AI in a way that really benefits them."

Anita Lettink

HR & Payroll Tech Advisor

Lano: If a business is determined to make its payroll and HR management smarter with the help of AI, how can they get there? What are the prerequisites?

Anita Lettink: First, you have to decide who you are going to position AI to. Is it for employees or is it for back-office functions to help people run HR and payroll? Then you also need to think about your processes. Are your HR and payroll processes the same everywhere or do they differ because of local requirements or preferences. Last but not least, you have to think about your data. What kind of data do you have? Where is it stored? Is the data reliable and of high quality? So, these are some of the things that organizations need to check before they can even start to think about using AI.

Lano: We always talk about AI using it as a universal term although there are many different types of AI so to speak, like digital assistants and chatbots. Which types of AI are actually being used in payroll and HR management and for what purpose?

Anita Lettink: The use of AI is not new in payroll and HR. Generative AI as a technology is new, but generally speaking, we have been using AI in payroll and HR for probably close to a decade now. Most Applicant Tracking Systems use artificial intelligence to quickly sort through high numbers of applicants to select the ones that are most appropriate for the role.

Lano: That’s an example from the HR space. Do you have some examples of AI use in payroll, too?

Anita Lettink: There are many use cases of AI in payroll. For instance, we've been using machine learning to analyze and verify the correctness of data and flag inconsistencies. We've also used robotic process automation, which maybe is not AI in the strictest sense of the word, but it is a form of automation, to optimize payroll processes in the backend. There are all kinds of applications of AI in HR and payroll. I think that one of the reasons why we're now talking so much about it is that generative AI became freely accessible for everybody and suddenly people understand the opportunity and the potential. But that does not mean that AI is new to payroll.

Lano: Could you describe the potential and possibilities that you see in using generative AI in payroll and HR?

Anita Lettink: The real potential of generative AI is that it can communicate in a way that is very natural to humans. It creates sentences and it can translate what you intend to do, no matter how we phrase it, which makes interaction between devices and people much easier. For instance, until recently, chatbots were scripted. So, if I asked a question, then there were four different answers and based on my input, the bot selected one of these answers. But if I did not phrase my question very well, then the answer I was looking for might not be on the list because the bot did not understand me. As a consequence, these interactions were never very satisfying to people. But the quality of this interaction can improve exponentially by using generative AI because it is much better at understanding what you mean.

"The real potential of generative AI is that it can communicate in a way that is very natural to humans."

Anita Lettink

HR & Payroll Tech Advisor

Lano: You mentioned that it's mainly about improving the quality of the exchange between people and devices. Do you have a concrete example of how this can help payroll and HR professionals in their daily tasks?

Anita Lettink: I don't think that the purpose of an HR professional is, for instance, to write job descriptions. Text composition is not their work, and while some of them are great at it, others struggle to put a good job description together. But what I do think is that these types of applications can help people craft great job descriptions more easily and faster. So they can focus on making sure that the people that are hired actually perform against that job description.

Lano: This is an example of how AI can help us perform better. But there is also a growing concern that AI will replace many jobs in the long term. What’s your take on this regarding payroll and HR?

Anita Lettink: I am trying to understand all these conversations about AI and how it will replace us. But my suggestion is that instead of focusing on the “will humans be replaced” question, we should think about how to leverage AI to help us perform better in our jobs and increase the quality of our work. All this debate on AI taking over and replacing people makes it scary when we should really focus on AI as a means to help us make our lives easier.

"Instead of focusing on the “will humans be replaced” question, we should think about how to leverage AI to help us perform better in our jobs."

Anita Lettink

HR & Payroll Tech Advisor

Lano: And to also be more productive at work.

Anita Lettink: There is a lot of debate about productivity. But in fact, all the great inventions and the technologies for digitizing work haven’t really enabled us to become more productive. Every new technology has created more work because it requires more people to manage that technology, but not necessarily making us more productive. So if we can find solutions and technologies to help us focus on work and perform that work with higher quality, higher accuracy, and higher speed, that would be great.

Lano: So, it’s ultimately just a question of knowing how to properly use AI to make our lives easier and gain some extra time. But where do you see the limits of AI?

Anita Lettink: Currently, it’s the case that people are responsible for their actions. So when something goes wrong, people are held liable. When using AI, it’s not possible to point at technology and blame it for any mistakes that might have happened. You need to be in control of what that technology does. It’s similar to working with a solution provider. If something goes wrong in payroll, for instance, you cannot point to your payroll provider and say that it was their mistake. The law says that you, as the employer, are responsible for making sure that your employees are being paid correctly and on time even if you outsource payroll to an external provider. This means that no matter where you add AI to your processes, you remain responsible if something goes wrong.

"No matter where you add AI to your processes, you remain responsible if something goes wrong."

Anita Lettink

HR & Payroll Tech Advisor

Lano: This is a very crucial point, especially when looking at the latest advances towards regulating the use of AI, namely the European AI Act.

Anita Lettink: Yes, the European AI Act is very specific about where you can apply AI and where you can't. It also highlights areas where you need to take additional precautions, and HR and payroll systems are considered high risk in the AI Act. This means that you must absolutely be in control. Yet the one thing that I don't see often enough is an open conversation between clients and tech vendors about what is going on within these solutions. I think that vendors are absolutely right to incorporate AI technology to make their solutions better. But I don't think that, at this point in time, clients understand enough of what is happening. Vendors need to start this conversation with their clients.

Lano: Yes, the inner workings of such tech solutions can resemble a black box. But how can vendors communicate more openly with their clients about how they use AI in their solutions?

Anita Lettink: For one thing, vendors need to make sure that they have some sort of official approach in place that they can share. And they need to have guidelines for using AI so that clients can be assured that, at the end of the day, it's always a person who makes the final decision.

Lano: Other than this lack of understanding and communication, what are the challenges businesses face if they want to implement AI in the workflows and processes?

Anita Lettink: Most companies are too small to run their own AI pilots, especially when it comes to HR. They don't have enough data and they don't have the budget. They are also not able to get the necessary talent because that talent is scarce and most of them are currently working at big tech companies. This partially explains why the average company is behind on governing AI.

Lano: That’s also the reason why so many organizations rely on external service providers, especially when it comes to tech solutions.

Anita Lettink: Yes, most companies depend on their vendors to run these solutions and infuse them with AI. Vendors typically have the budget and the data to run all kinds of pilots. But, again, what I don't see enough is that they communicate with their clients about their pilot programs, the results, and the implementation in their solutions. I think that vendors across the world, not just in HR and payroll but in general, need to be much more open about how they use data, what they do with it, how they apply AI, what the outcomes are, and how it is beneficial. But also where the risks are.

"Vendors need to be much more open about how they use data, what they do with it, and how they apply AI to it."

Anita Lettink

HR & Payroll Tech Advisor

Lano: That’s an important point to mention. Using AI is never without risks. But especially in areas like HR and payroll, vendors and businesses need to take extra precautions.

Anita Lettink: Yes, there are obligations on both the client and the vendor side. Vendors cannot just tell their clients that they incorporated AI into their systems and expect them to trust their solution blindly. They need to reach out to their clients and inform them about what they are doing and what the involved risks and benefits are. But there is also an obligation on the client to educate themselves and start having tough conversations with their vendors. There is a mutual responsibility to have an in-depth conversation on this topic.

Lano: How would you rate the readiness of organizations to take this active approach?

Anita Lettink: I think that we're halfway there. But in the current situation where AI is suddenly everywhere, we really need to think this through and make decisions. And especially in Europe, there is another important aspect that needs to be considered, and that is the employees. With the strong role of works councils and employee representation committees, employees need to be taken into consideration when selecting new HR and payroll solutions. They need to be involved, and they are bound to bring up the topic of AI use if vendors and employers are not on top of this.

Lano: So far, we’ve only covered the vendor and the business perspective, but what about employees? When we look at it from the employee perspective, what is the potential impact AI can have on employees?

Anita Lettink: I think employees could look at AI as a personal assistant or personal coach. For instance, in the past, employees had to go through HR or maybe talk to an HR business partner to understand what career options they had with their qualifications, skills, and current job role. Just to get some information and make a decision. Today, employees could potentially ask an AI assistant to bring up that information for them. In this case, the AI could give an outlook of potential opportunities in the company in a year or two based on the requester’s data. So it will be a much more personalized, much more tailored experience compared to the sort of generic profile that you would get in the past.

Lano: Do you have another example of where AI can come in handy for employees?

Anita Lettink: You could also look at AI as a personal job coach that helps employees to quickly get answers and guides them through the tasks that they need to accomplish. We know from research that AI assistants are especially good at helping less experienced employees get up to speed quickly. Very skilled workers are usually more distracted by AI suggestions because they have a much better understanding of what works and what doesn't, and they can finish that work intuitively. But I wouldn’t underestimate the power of using AI in the first few months or years to build on existing knowledge and get help to perform according to the job requirements.

Lano: Going back to the corporate side, where do you see the main gains and benefits AI holds for businesses regarding payroll and HR?

Anita Lettink: I would say the focus is on helping HR professionals and payroll admins do their job quicker and with higher quality. I absolutely see potential to improve these back-office processes with AI. We've been able to do anomaly detection for a very long time, meaning that AI is flagging problems for payroll admins to look at. What you can do now is to also let AI look at the solutions suggested by the payroll professional and learn from it to suggest an appropriate solution. This helps payroll admins move from a very operational, tactical role to a more strategic one. That’s simply because once they start to have command of the data and can use AI to analyze that data more quickly, they can start to really focus on delivering business support. We need to automate and use AI for everything we can from a payroll perspective.

"A payroll professional always needs to be in control and always needs to be the final person to make the decision."

Anita Lettink

HR & Payroll Tech Advisor

Lano: So, automation and AI from start to finish?

Anita Lettink: I don't mean front to back. I mean where that makes sense. A payroll professional always needs to be in control and always needs to be the final person to make the decision. But I do think that there's a lot more that we can do in an automated way. For instance, you could use AI to keep track of all the legislation that is changing on a daily basis. AI could read and process new legislation in the background and then summarize the main points that have changed compared to the current solution set-up. What should not happen, however, is that AI makes those adjustments automatically in the background, with the payroll professional only seeing the changes when they are actually implemented on payslips. That is not what you want.

Lano: To sum this up, when you use AI, you need to be responsible for it and, more than anything, you need to be in control. Now, looking at the future, what's next on the horizon when it comes to AI in payroll and HR?

Anita Lettink: Last year, we saw a lot of companies that announced their new AI features. This year, we don't see that many announcements. I just recently wrote a newsletter where I raised the question of whether we needed better chatbots or just better service? What I'm advocating is to be careful in assuming that AI is the answer to all your problems. Before using AI, you need to ask yourself what the problem is and what the best technology is to solve that problem. And when you do that, maybe you come to the conclusion that AI will really help you in certain parts. But in other parts, you might not need it to deliver a better service. I hope that over the next one or two years, we will see an increase of AI adoption in all kinds of processes, but I also hope that it will come with a healthy dose of realism and that we step away from the hype to really make it work for us. Because there is an opportunity here, but it's not that AI will do everything.

Lano: That’s the perfect conclusion to end the interview. Thank you so much for this insightful conversation, Anita.

"What I'm advocating is to be careful in assuming that AI is the answer to all your problems."

Anita Lettink

HR & Payroll Tech Advisor

Anita Lettink is an international keynote speaker, advisor and author. After a global career in HR and payroll consulting and outsourcing, she started her own business to help CHROs and their teams understand how work is changing and what they can do to prepare their employees and companies to thrive.

Anita has been recognized as a Top 25 Global Thought Leader on the Future of Work and a Top 100 HR Tech Influencer. She is the author of “How to Select Your Next Payroll” and “Equal Pay for Equal Work”. Anita is Partner at Strategic Management Centre and the founder of HRTech Radar.