Microsoft Study: Generative AI's Impact on Jobs Revealed

A recent study by Microsoft Research provides new insights into how generative AI is impacting various occupations, identifying jobs most and least susceptible to its influence. The study, which analyzed 200,000 anonymized conversations with Bing Copilot users in 2024, offers a unique, real-world perspective on AI's integration into the workplace.

Knowledge Work and Communication Most Impacted

The research indicates that occupations heavily reliant on knowledge work and communication are most vulnerable to generative AI. This includes roles such as interpreters and translators, historians, writers and authors, sales representatives, and customer service representatives. Tasks involving frequent writing, information retrieval, editing, and client communication are areas where AI demonstrates significant applicability. Other highly impacted roles include technical writers, editors, public relations specialists, mathematicians, and data scientists. The study found that generative AI is particularly adept at tasks like information gathering, text generation and editing, and communication and explanation. Notably, some of these highly affected jobs, such as journalists and library science professors, often require a bachelor's degree or higher, challenging the traditional belief that higher education guarantees job security in an evolving workforce.

Manual Labor and Physical Presence Remain Least Affected

Conversely, jobs involving manual labor, physical presence, or machinery operation ranked lowest in terms of AI applicability. Professions least likely to be impacted by generative AI include dredge operators, bridge and lock tenders, water treatment plant and systems operators, rail track maintenance workers, roofers, and massage therapists. Other roles less affected are nursing assistants, dishwashers, and industrial truck and tractor operators. These jobs often require physical movement, hands-on equipment handling, or direct human interaction, areas where current text-based AI tools have limited functional roles.

Methodology and Broader Implications

Microsoft's study is distinct because it analyzed actual AI usage data from Bing Copilot conversations, rather than relying solely on theoretical assessments. Researchers mapped these conversations to work activities using the U.S. government's job classification system (O*NET) and calculated an "AI applicability score" for each occupation. The score considers the AI model's success rate, the user's goal, and the task success rate.

It is important to note that the study emphasizes that a high "AI applicability score" does not necessarily equate to job displacement. Instead, generative AI tools are expected to augment workflows, leading to the creation of hybrid roles where human judgment and oversight remain crucial. While some companies, like IBM, have paused hiring for roles they anticipate AI will fill, the broader consensus, including from the World Economic Forum, suggests that AI will create more jobs than it displaces by 2030. The World Economic Forum's "Future of Jobs Report 2025" anticipates a net increase of 78 million jobs by 2030, with 170 million new roles created and 92 million displaced. This highlights a significant shift towards roles requiring technological literacy and new skills.

The rapid adoption of generative AI is already transforming business models and driving a surge in demand for GenAI training and upskilling. Companies are increasingly prioritizing reskilling and upskilling their workforce to enhance collaboration with AI systems. The evolving landscape underscores the need for continuous learning and adaptation, as even entry-level roles in some white-collar professions are being redefined to require proficiency with generative AI tools.