Would you trust AI to decide your pay rise?

While algorithms promise faster and smarter human resource management, employees may see the process as colder and less human

Featured faculty: Choi Sungwoo
Written by Pan Jingyi

Derek Mobley had applied to more than 100 jobs over several years and had admittedly received rejection notices within minutes or hours. Confused and furious, the US citizen sued Workday, a human resources software firm that handled most of his applications, alleging that its artificial intelligence (AI) screened out his applications based on his age, race, and disabilities.

While Workday has argued that it’s not liable for hiring decisions, a court conditionally certified the age discrimination claims last year. This highly anticipated lawsuit will set a new precedent for AI-driven hiring practices and serves as a reminder that handing over employment decisions to algorithms can lead to backlash.

AI has inevitably reshaped human resource management, as also seen across industries today. A 2025 survey from Resume Builder shows that a majority of US managers have relied on AI for high-stakes decisions, such as promotions, rises and even layoffs. Akin to a dystopian Black Mirror episode, bosses are turning to machines to decide who’s in and who’s out.

Although more and more companies embed automation into their human resource operations, little is known about the impact on employee morale when their career is in the hands of a non-human actor. This is the puzzle that Choi Sungwoo, Assistant Professor of the School of Hotel and Tourism Management at the Chinese University of Hong Kong (CUHK) Business School, seeks to answer.

Professor Choi uncovers a latent repercussion of advanced technology in human resource management: organisational dehumanisation. “Organisational dehumanisation is the feeling of being reduced to a mere functional component of an organisation, much like a single bolt in a large machine, where your unique qualities, emotions and individuality are largely disregarded.”

Why AI can feel dehumanising

In a study titled AI in human resource management: A driver of organisational dehumanisation and negative employee reactions, Professor Choi works with Shin Hyejo of the Hong Kong Polytechnic University and Kim Hyunsu of the University of Macau on three scenario-based online experiments. They recruited nearly 700 participants through Prolific, an online platform widely used in academic research.

“When AI performs human resources operations, employee characteristics are seen as numbers. Therefore, employees would feel like they are not treated as humans.”

Professor Choi Sungwoo

Across all three experiments, participants were asked to imagine that decisions about their promotion or performance review were being made either by an AI system or by a human manager. The results were consistent: people reacted more negatively when the decision-maker was AI.

Those reactions were not trivial. Participants reported lower commitment, stronger turnover intentions, and even greater retaliatory feelings when seeing AI deciding their livelihood. In practice, they are more likely to search for another job and warn others to avoid working for the company.

A couple of factors drive such dehumanising feelings, Professor Choi notes. AI lacks the ability to understand social norms, personal issues and ethical concerns as a human manager can. AI also works in incomprehensible ways to laypeople, and employees may fail to understand how AI reaches its conclusions. As a result, they feel powerless and excluded from the decision-making process.

“Putting it all together, loss of empathy, transparency, and control can leave people feeling objectified,” Professor Choi says. “When AI performs human resources operations, employee characteristics are seen as numbers. Therefore, employees would feel like they are not treated as humans.”

Can human resources automation thrive?

Different companies have different sentiments towards AI. Professor Choi and his collaborators identify what they describe as a cultural paradox: companies with more collaborative and family-like cultures may experience greater resistance to AI in human resources management.

In these collaborative environments, employees believe that the management values cooperation, support and interpersonal relationships. If AI is then used to make major decisions, the technology can clash with such a principle.

“When a collegial workplace adopts AI for major human resource decisions, it can feel like a betrayal,” Professor Choi explains. “Employees experience a dissonance between the human-centred values the culture espouses and the perceived quantification of their worth that AI involvement implies.”

By contrast, the negative effects of AI appear to be less intense in companies with a more outcome-oriented system, where performance and results are already prioritised over interpersonal connections. But that does not mean AI is entirely harmless in such settings.

Even in outcome-focused organisations, AI-driven decisions on the workforce can still backfire and brew dehumanisation. Companies that deploy an AI system to select the most suitable candidates for promotion still see employees feel their aspirations or contributions are overlooked.

In short, whether focusing on outcomes or collaborations, the company needs to take the human aspect into account when adopting AI in human resources operations.

Keep “human” in human resources

Reputation is a valuable asset for a company. When job marketplaces like Glassdoor, Indeed, Seek, and even Google nowadays provide user-generated company reviews, the efficiency gains from AI might be quickly diminished by a wave of criticism from current and former employees.

For companies and business leaders, the message is not to abandon AI, but to use it wisely. One priority, Professor Choi says, is transparent communication. “Companies should clearly explain why AI adoption in human resources is necessary and reassure employees that it does not compromise the organisation’s core commitment to supportive and human-oriented growth.”

RELATED ARTICLE

How assistive robots can boost an inclusive workforce

Rather than allowing AI to make decisions on its own, Professor Choi suggests designing systems that enable humans and technology to coexist. “This hybrid approach helps preserve the sense that consequential decisions about people are ultimately made by people. If AI serves only in an assistive capacity with limited input into the final decision, the dehumanisation effect should be substantially mitigated.”

“Human resource practices are highly sensitive, and AI could be most valuable in managing less critical yet voluminous tasks, such as initial application screening, freeing human managers to focus on more complex and high-stakes decisions,” Professor Choi adds. “Despite that, companies should always be mindful of the risks of dehumanising feelings and act accordingly.”