Explore the rankings
Science & Research · Updated Aug 2026

Physicists

Conduct research into physical phenomena, develop theories on the basis of observation and experiments, and devise methods to apply physical laws and theories.

JVS 2.0.0-phase4b
AI Exposure
74/100
High

How much of this occupation’s work can be materially affected by current AI systems.

Replacement Risk
69/100
High

How likely exposure is to translate into reduced human demand.

Includes provisional estimates for AI adoption pressure and labour-market resilience. How this is measured

Evidence quality
Confidence82/100
Task coverage84%

Confidence reflects task coverage, mapping and capability-evidence quality, and how much of the score rests on provisional inputs.

Task-level evidence

What is driving the score?

Occupation scores are built from the task mix—not a single prediction about a job title.

JVS 2.0.0-phase4b
TaskImportanceAI impactExposure
Perform complex calculations as part of the analysis and evaluation of data, using computers.High
80
Describe and express observations and conclusions in mathematical terms.High
82
Analyze data from research conducted to detect and measure physical phenomena.High
68
Design computer simulations to model physical data so that it can be better understood.High
80
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.Medium
79
Collaborate with other scientists in the design, development, and testing of experimental, industrial, or medical equipment, instrumentation, and procedures.Medium
80
Report experimental results by writing papers for scientific journals or by presenting information at scientific conferences.Medium
80
Observe the structure and properties of matter, and the transformation and propagation of energy, using equipment such as masers, lasers, and telescopes, to explore and identify the basic principles governing these phenomena.Medium
37
Most exposed

Where AI can do more

Routine, digitized, and highly repeatable tasks face the greatest pressure.

  1. Describe and express observations and conclusions in mathematical terms.82
  2. Perform complex calculations as part of the analysis and evaluation of data, using computers.80
  3. Design computer simulations to model physical data so that it can be better understood.80
  4. Collaborate with other scientists in the design, development, and testing of experimental, industrial, or medical equipment, instrumentation, and procedures.80
Hardest to automate

Where people still matter

These tasks score lowest on automation feasibility—physical presence, judgement, accountability and real-world variability all resist end-to-end automation.

  1. Observe the structure and properties of matter, and the transformation and propagation of energy, using equipment such as masers, lasers, and telescopes, to explore and identify the basic principles governing these phenomena.01
  2. Analyze data from research conducted to detect and measure physical phenomena.02
  3. Perform complex calculations as part of the analysis and evaluation of data, using computers.03
  4. 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.04
  5. Collaborate with other scientists in the design, development, and testing of experimental, industrial, or medical equipment, instrumentation, and procedures.05
Where else this work leads

Related occupations

Occupations O*NET links to this one. Relatedness reflects shared work, not a claim that these roles are safer.

See all rankings →
Beyond AI capability

Adoption and labour-market outlook

Structural factors are kept separate from raw capability so you can see what actually resists automation. Adoption pressure and labour-market resilience are still provisional models—25% of this occupation’s replacement-risk weight rests on them.

Human dependency55
Physical dependency17
Adoption pressure57
Labour-market resilience50
Methodology & sources

O*NET 30.3 occupational data interpreted through the JobsVsAI capability, automation and structural-constraint models.

Confidence82/100
CalculatedAug 21, 2026
Read methodology →