Harsh Agarwal
Research Subjects
Research Interests
Harsh Agarwal has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Akshat Rana
Netaji Subhas University of Technology , Delhi
1 shared publication - Avinash Agarwal
Telecommunication Engineering Centre Ministry of Communications New Delhi India
1 shared publication Nihaarika Agarwal
Manipal Institute of Technology, Manipal, India
1 shared publication- Peeyush Agarwal
Netaji Subhas University of Technology , Delhi
1 shared publication
Latest Publications
- 29 September 2026
- Published by Emerald Publishing in International Journal of Innovation Science
- p. 1-23
What work is AI actually doing? Uncovering the drivers of generative AI adoption
Purpose: The rapid integration of artificial intelligence (AI) systems such as ChatGPT, Gemini and Claude AI is reshaping how work is done. Predicting how AI will reshape work requires evidence on how the technology is actually being adopted to complement existing evidence on its capabilities. This study aims to investigate which intrinsic task characteristics are associated with users’ decisions to delegate work to AI systems. Design/methodology/approach: This study uses the Anthropic Economic Index data set of four million Claude AI interactions mapped to O*NET tasks. The authors systematically scored each task across seven characteristics: routine, cognitive, social intelligence, creativity, domain knowledge, complexity and decision making, operationalized through 35 underlying parameters. The authors then used multivariate techniques to identify latent task archetypes and analyzed their relationship with AI usage. Findings: Tasks requiring high creativity, complexity and cognitive demand, but low routineness, attracted the most AI engagement. The authors identified three task archetypes: Dynamic Problem Solving, Procedural and Analytical Work and Standardized Operational Tasks, showing that AI usage is more strongly associated with combinations of task characteristics than with individual factors alone. The analysis revealed highly concentrated AI usage patterns, with just 5% of tasks accounting for 59% of all interactions. Originality/value: This research links observed task-level AI adoption to a multi-dimensional framework of intrinsic task characteristics. It provides the first systematic, characteristics-level evidence to complement occupation-level exposure estimates. It also introduces a data-driven classification of work archetypes that offers a new lens for analyzing the human-AI division of labor.
- 29 September 2026
- Published by Emerald Publishing in International Journal of Innovation Science
- p. 1-23
What work is AI actually doing? Uncovering the drivers of generative AI adoption
Purpose: The rapid integration of artificial intelligence (AI) systems such as ChatGPT, Gemini and Claude AI is reshaping how work is done. Predicting how AI will reshape work requires evidence on how the technology is actually being adopted to complement existing evidence on its capabilities. This study aims to investigate which intrinsic task characteristics are associated with users’ decisions to delegate work to AI systems. Design/methodology/approach: This study uses the Anthropic Economic Index data set of four million Claude AI interactions mapped to O*NET tasks. The authors systematically scored each task across seven characteristics: routine, cognitive, social intelligence, creativity, domain knowledge, complexity and decision making, operationalized through 35 underlying parameters. The authors then used multivariate techniques to identify latent task archetypes and analyzed their relationship with AI usage. Findings: Tasks requiring high creativity, complexity and cognitive demand, but low routineness, attracted the most AI engagement. The authors identified three task archetypes: Dynamic Problem Solving, Procedural and Analytical Work and Standardized Operational Tasks, showing that AI usage is more strongly associated with combinations of task characteristics than with individual factors alone. The analysis revealed highly concentrated AI usage patterns, with just 5% of tasks accounting for 59% of all interactions. Originality/value: This research links observed task-level AI adoption to a multi-dimensional framework of intrinsic task characteristics. It provides the first systematic, characteristics-level evidence to complement occupation-level exposure estimates. It also introduces a data-driven classification of work archetypes that offers a new lens for analyzing the human-AI division of labor.
- 7 March 2022
- Published by Springer Nature in AI and Ethics
- Vol. 3 (1) , 267-279
Fairness Score and process standardization: framework for fairness certification in artificial intelligence systems
- 7 March 2022
- Published by Springer Nature in AI and Ethics
- Vol. 3 (1) , 267-279
Fairness Score and process standardization: framework for fairness certification in artificial intelligence systems