As artificial intelligence (AI) rapidly integrates into workplaces, fears of job displacement are growing among workers and new entrants to the labor market. Many are searching for “AI-proof” careers to secure their future as companies increasingly opt for automation over hiring. While no study can definitively predict which jobs are immune to AI, a recent Microsoft study offers insight into which roles are least and most likely to be affected by automation, based on real-world data.
Microsoft’s Study: Measuring AI Applicability
Published in August 2025, Microsoft’s study analyzed over 200,000 anonymized conversations with its Bing Copilot chatbot from January to September 2024. The goal was to understand how people use generative AI for work tasks and to what extent these tasks align with specific occupations. The researchers developed “AI applicability scores” by evaluating three factors:
- Coverage: How often people use AI for specific job tasks.
- Completion Rate: How successfully AI completes those tasks.
- Impact Scope: The extent to which AI can assist or perform the work.
These scores highlight where AI is most and least effective in current workplace tasks, offering a snapshot of automation’s potential impact.
Important Caveats
Microsoft emphasizes that high AI applicability doesn’t mean jobs will vanish. “Our research shows AI supports tasks, particularly in research, writing, and communication, but it doesn’t fully perform any single occupation,” the researchers told Gizmodo. The study focuses on task overlap, not job replacement, and doesn’t account for downstream economic effects, such as wage changes or new roles created by AI. The data is also U.S.-centric and based solely on Bing Copilot, so other AI tools or global trends may differ.
Why Companies Are Turning to AI
Many executives see AI as a tool to cut labor costs and boost profits. Ford CEO Jim Farley recently predicted at the 2025 Aspen Ideas Festival that AI could replace half of U.S. white-collar workers. Some companies have already implemented policies requiring managers to justify why a role can’t be automated before hiring. This trend has slowed hiring, particularly for early-career white-collar workers, who face the highest risk of AI disruption, according to Microsoft’s findings.
However, automation isn’t always the best choice. While AI excels in tasks like writing or data analysis, its outputs—such as AI-generated text—face criticism for issues like copyright violations and lack of originality. Former Google executive Mo Gawdat warned in August 2025 that unchecked AI automation could lead to a “short-term dystopia” within 15 years, disrupting labor markets and exacerbating inequality. Experts, including Microsoft’s researchers, advocate for AI augmentation—using AI to enhance human work—over full automation to balance productivity gains with job stability.
Jobs Least Likely to Be Automated
Microsoft’s study identifies roles with low AI applicability, meaning they rely on physical skills, human interaction, or specialized expertise that AI struggles to replicate. These “AI-proof” jobs are predominantly in healthcare and blue-collar fields, where datasets for AI training are limited (e.g., less than 10% of surgical data is publicly available due to regulations). The top 10 jobs with the lowest AI applicability scores are:
- Phlebotomists
- Nursing Assistants
- Hazardous Materials Removal Workers
- Helpers (Painters, Plasterers)
- Embalmers
- Plant and System Operators
- Oral and Maxillofacial Surgeons
- Automotive Glass Installers and Repairers
- Ship Engineers
- Tire Repairers and Changers
These roles require hands-on expertise or physical presence, making them less susceptible to generative AI, which excels in data-driven tasks.
Jobs Most Vulnerable to AI
Conversely, the study highlights jobs with high AI applicability, primarily knowledge-based and communication-heavy roles where AI is already being integrated. The top 10 jobs with the highest AI applicability scores are:
- Interpreters and Translators
- Historians
- Passenger Attendants
- Sales Representatives of Services
- Writers and Authors
- Customer Service Representatives
- CNC Tool Programmers
- Telephone Operators
- Ticket Agents and Travel Clerks
- Broadcast Announcers and Radio DJs
These roles involve tasks like writing, research, and customer interaction, where AI chatbots like Copilot perform well. However, high applicability doesn’t guarantee replacement—AI often augments these tasks, acting as a coach or assistant rather than a full substitute.
Navigating the AI-Driven Workplace
The Microsoft study underscores a key reality: AI is reshaping how work is done, not necessarily eliminating jobs outright. Workers in high-risk fields can stay competitive by focusing on uniquely human skills—creativity, empathy, and complex problem-solving—that AI can’t replicate. Upskilling in areas like AI oversight or hybrid workflows can also help. For companies, the challenge is balancing cost-cutting with the risks of over-automation, which could compromise quality or innovation.
As AI continues to evolve, both workers and employers must adapt. By understanding which tasks AI can enhance and which remain human-driven, individuals can make informed career choices, and companies can integrate AI thoughtfully to boost productivity without destabilizing the workforce.