BETA
Updated: Mar 24
BETA
Updated: Mar 24
Evolving
Last Update: 3/13/2026
Your role’s AI Resilience Score is
45.6%
Median Score
Changing Fast
Evolving
Stable
This reflects the reliability of your score based on the number of data sources available for this career and how closely those sources agree on the outlook. A higher confidence means more consistent evidence from labor experts and AI models.
What does this resilience result mean?
These roles are shifting as AI becomes part of everyday workflows. Expect new responsibilities and new opportunities.
AI Resilience Report for
Physicists
They study how the universe works by exploring the laws of nature, conducting experiments, and applying their findings to solve real-world problems.
This role is evolving
The career of a physicist is labeled as "Evolving" because AI is increasingly used to handle routine tasks like data analysis, which speeds up research processes. While AI can help with complex calculations and finding patterns, human physicists are still essential for interpreting results, designing experiments, and writing research papers.
Read full analysisLearn more about how you can thrive in this position
View analysis
Chat with Coach
Latest news
More career infoAnalysisChatNewsMoreLearn more about how you can thrive in this position
View analysisChat with CoachLatest newsMore career infoAnalysisChatNewsMoreThis role is evolving
The career of a physicist is labeled as "Evolving" because AI is increasingly used to handle routine tasks like data analysis, which speeds up research processes. While AI can help with complex calculations and finding patterns, human physicists are still essential for interpreting results, designing experiments, and writing research papers.
Read full analysisContributing Sources
We aggregate scores from multiple models and supplement with employment projections for a more accurate picture of this occupation’s resilience. Expand to view all sources.
AI Resilience
AI Resilience Model v1.0
AI Task Resilience
CareerVillage's proprietary model that estimates how resilient each occupation's tasks are to AI automation and augmentation
Evolving48.0%
48.0%
Microsoft's Working with AI
AI Applicability
Measures how applicable AI tools (like Bing Copilot) are to each occupation based on real usage patterns
Changing fast24.9%
24.9%
Anthropic's Observed Exposure
AI Resilience
Based on observed patterns of how Claude is being used across occupational tasks in real conversations
Changing fast20.9%
20.9%
Will Robots Take My Job
Automation Resilience
Estimates the probability of automation for each occupation based on research from Oxford University and other academic sources
Stable95.4%
95.4%
Althoff & Reichardt
Economic Growth
Measured as "Wage bill" which is a long term projection for average wage × employment. It's the total labor income flowing to an occupation
Evolving36.8%
36.8%
Medium Demand
Labor Market Outlook
We use BLS employment projections to complement the AI-focused assessments from other sources.
Learn about this scoreGrowth Rate (2024-34):
4.0%
Growth Percentile:
62.9%
Annual Openings:
1,700
Annual Openings Pct:
18.9%
Analysis of Current AI Resilience
Physicists
Updated Quarterly • Last Update: 2/17/2026
What's changing and what's not
Physicists already use powerful computers and software for much of their routine work – for example, specialized programs like Mathematica or MATLAB handle the hard math and simulations [1]. In recent years, AI methods (especially machine learning) have begun to help with data-heavy tasks. For instance, an AI “Virtual Research Assistant” at Oxford University was able to scan telescope data and flag supernova explosions, cutting the manual data-sifting work by about 85% [2].
This shows that one core task – analyzing research data to detect phenomena – can be greatly sped up by AI. Likewise, research reviews note that machine learning is emerging across many physics fields and could even “revolutionize” how we understand complex data [3].
However, not all physics tasks are automated today. Tasks like writing research papers or collaborating on experiments still rely on human insight [1]. Scientists must interpret results, design experiments, and explain findings in creative ways, and current AI tools cannot replace that.
Even in the examples above, humans had to set up the AI models and review their output. As one report notes, advanced AI methods for labs (like large language model assistants) exist only in places with already automated equipment [4]. In short, computers and AI are taking over more of the number-crunching and pattern-finding in physics, but human physicists still do the high-level thinking, planning, and writing.
Sources
AI in the real world
The speed of AI adoption in physics depends on many factors. On one hand, the economic benefits can be large. The Oxford supernova assistant, for example, was built with a small AI model running on a laptop—so it didn’t need a huge supercomputer—and it saved the team a lot of time [5] [2].
This kind of success encourages more use of AI: if an AI tool cuts our work by half or more, researchers will pay attention. On the other hand, building AI for physics can be hard and costly. It often requires expert knowledge of both physics and computer science, and many physics labs are not yet set up for it [4] [3].
For example, one study found that AI-driven lab helpers are mostly used in a few advanced labs that already have automated instruments [4].
Social and ethical trust also matter. Physicists are careful people – they want to double-check AI results. In the Oxford case, scientists still verify every candidate supernova, meaning human “sign-off” is always needed [2] [2].
In general, because physics research often involves complex, important problems, the community is willing to try AI only where it clearly helps and with humans in charge. In summary, AI tools for physicists are steadily growing: they speed up calculations and data analysis and let researchers focus on big ideas, but they are used as helpers, not replacements. Young physicists can be reassured that their creativity and insight remain vital – AI is a tool to make their work easier, not to do the science for them [2] [3].
Sources
Help us improve this report.
Tell us if this analysis feels accurate or we missed something.
Share your feedback
More Career Info
Career: Physicists
Parent Careers
Similar Careers
Employment & Wage Data
Median Wage
$166,290
Jobs (2024)
24,600
Growth (2024-34)
+4.0%
Annual Openings
1,700
Education
Doctoral or professional degree
Experience
None
Source: Bureau of Labor Statistics, Employment Projections 2024-2034
Task-Level AI Resilience Scores
AI-generated estimates of task resilience over the next 3 years
1
Develop manufacturing, assembly, and fabrication processes of lasers, masers, infrared, and other light-emitting and light-sensitive devices.
2
Teach physics to students.
3
Develop theories and laws on the basis of observation and experiments, and apply these theories and laws to problems in areas such as nuclear energy, optics, and aerospace technology.
4
Direct testing and monitoring of contamination of radioactive equipment, and recording of personnel and plant area radiation exposure data.
5
Advise authorities of procedures to be followed in radiation incidents or hazards, and assist in civil defense planning.
6
Report experimental results by writing papers for scientific journals or by presenting information at scientific conferences.
7
Develop standards of permissible concentrations of radioisotopes in liquids and gases.
Tasks are ranked by their AI resilience, with the most resilient tasks shown first. Core tasks are essential functions of this occupation, while supplemental tasks provide additional context.
© 2026 CareerVillage.org. All rights reserved.
The AI Resilience Report is a project from CareerVillage.org®, a registered 501(c)(3) nonprofit.
Built with ❤️ by Sandbox Web
The AI Resilience Report is governed by CareerVillage.org’s Privacy Policy and Terms of Service. This site is not affiliated with Anthropic, Microsoft, or any other data provider and doesn't necessarily represent their viewpoints. This site is being actively updated, and may sometimes contain errors or require improvement in wording or data. To report an error or request a change, please contact air@careervillage.org.