AI and Human-Job Matching: What Makes Humans Succeed in AI-Augmented Work (AI-MATCH)

AI-MATCH explores why some people thrive with AI tools while others struggle, even with similar backgrounds.

Through large-scale experiments with UCL students, the project aims to uncover who benefits from AI and why, shaping education, workforce training, and policy for an equitable future.

Artificial intelligence is rapidly transforming workplaces, but we don't fully understand who benefits and who gets left behind. While some people use AI tools to dramatically boost their productivity, others struggle to gain any advantage. This variation exists even among people with similar educational backgrounds. Our project, AI-MATCH, investigates a critical question: What determines who succeeds when working with AI? 

Consider two university students - one studying English literature, the other computer science. When given AI assistance, might the English student suddenly be able to write computer code? Or does AI simply make the computer science student even more advantaged? These aren't just academic questions - they have profound implications for education policy, workforce training, and social equality. 

Our research brings together expertise from computer science and economics to conduct a large-scale experiment with 400 UCL students. We're designing realistic workplace tasks - from coding challenges to business analyses - that mirror what professionals actually do. Students will complete cognitive and personality assessments, then tackle these tasks both with and without AI assistance. We'll measure not just whether they succeed, but how they interact with AI: their conversation patterns, prompting strategies, and problem-solving approaches. 

This goes beyond asking "does AI help?" to understanding precisely who it helps and why. We're examining whether AI democratizes expertise (enabling non-specialists to perform specialized work) or concentrates it (amplifying existing advantages). We're also investigating how personality traits like openness to experience or conscientiousness influence AI collaboration success. 

The implications are far-reaching. If we discover that verbal skills become more valuable while technical skills matter less, this fundamentally changes what universities should teach. If certain personality traits predict AI success, we can develop targeted training programs. Our findings will help ensure AI's benefits are shared broadly rather than concentrated among those already advantaged. 

We're partnering with local and international organisations to provide global workforce development guidance. This matters because different countries have built their economies around different capabilities - India's coding expertise, the Philippines' customer service excellence, London's financial services. As AI transforms these sectors, entire nations need evidence-based strategies for adaptation. 

This pilot study lays the groundwork for understanding how AI will reshape not just individual careers but entire organizations. Future work will examine how AI changes team dynamics - when some team members excel at AI collaboration while others struggle, how do new workplace hierarchies emerge? Is AI becoming a new type of teammate requiring different coordination skills? 

By identifying where AI creates opportunities versus barriers, we enable targeted interventions: curriculum reforms ensuring all students develop AI-complementary skills; training programs teaching effective AI interaction techniques; and policy frameworks ensuring AI's transformative benefits reach everyone, not just those with certain backgrounds or personality traits. 

The urgency is clear. Students entering university today will graduate into a fundamentally different job market. Organizations are implementing AI tools without understanding their differential impacts on employees. Countries are making workforce development decisions with limited evidence. AI-MATCH provides the rigorous evidence needed to ensure AI enhances human potential broadly rather than exacerbating existing inequalities. 

Through this research, we're working toward a future where AI amplifies everyone's capabilities, where education prepares all students for AI collaboration, and where the benefits of this technological revolution are shared equitably across society.