Will soil scientists be replaced by AI?
Updated
Soil scientists score 41 of 100 for AI exposure. The largest share of their task time, 98%, is work AI speeds up while they stay in charge. The other 2% stays with people.
- 98% AI helps
- 2% Needs a person
Moderate exposure
Higher than 55% of the 923 occupations we scored.
Task by task
We graded 27 tasks, listed here from most to least exposed. AI can't do any of them end to end yet. It speeds up 26 and leaves 1 to people.
| Task | AI exposure |
|---|---|
| Communicate research or project results to other professionals or the public or teach related courses, seminars, or workshops. AI drafts papers, slides, and course materials; delivery stays human. | 50% AI helps |
| Provide information or recommendations to farmers or other landowners regarding ways in which they can best use land, promote plant growth, or avoid or correct problems such as erosion. AI can draft farmer advice; site-specific judgment stays with the scientist. | 40% AI helps |
| Identify or classify species of insects or allied forms, such as mites or spiders. AI image tools can aid insect ID; specimen handling stays manual. | 40% AI helps |
| Provide advice regarding the development of regulatory standards for land reclamation or soil conservation. AI can draft regulatory text and review literature; judgment stays human. | 35% AI helps |
| Study ways to improve agricultural sustainability, such as the use of new methods of composting. Literature review and modeling of sustainability practices are AI-assisted. | 35% AI helps |
| Study insect distribution or habitat and recommend methods to prevent importation or spread of injurious species. AI helps model insect distribution and draft prevention recommendations. | 35% AI helps |
| Develop methods of conserving or managing soil that can be applied by farmers or forestry companies. AI helps draft conservation plans; field validation remains human. | 30% AI helps |
| Identify degraded or contaminated soils and develop plans to improve their chemical, biological, or physical characteristics. AI can help model soil degradation and draft remediation plans. | 30% AI helps |
| Study soil characteristics to classify soils on the basis of factors such as geographic location, landscape position, or soil properties. AI assists soil classification from data; field description stays human. | 30% AI helps |
| Survey undisturbed or disturbed lands for classification, inventory, mapping, environmental impact assessments, environmental protection planning, conservation planning, or reclamation planning. Field surveys are physical; AI aids GIS mapping and report drafting. | 30% AI helps |
Show the other 17 tasks
| Task | AI exposure |
|---|---|
| Plan or supervise waste management programs for composting or farming. AI can draft compost program plans; site supervision stays human. | 30% AI helps |
| Research technical requirements or environmental impacts of urban green spaces, such as green roof installations. AI helps research technical specs and impacts; site work stays physical. | 30% AI helps |
| Investigate soil problems or poor water quality to determine sources and effects. Sampling is physical; AI helps interpret data and trace sources. | 25% AI helps |
| Investigate responses of soils to specific management practices to determine the use capabilities of soils and the effects of alternative practices on soil productivity. Field trials are hands-on; AI speeds data analysis and reporting. | 25% AI helps |
| Conduct experiments to investigate the underlying mechanisms of plant growth and response to the environment. Plant growth experiments are lab and field based; AI aids analysis. | 25% AI helps |
| Develop new or improved methods or products for controlling or eliminating weeds, crop diseases, or insect pests. Control method testing is physical; AI helps screen compounds and data. | 25% AI helps |
| Develop environmentally safe methods or products for controlling or eliminating weeds, crop diseases, or pests. Safe pest control testing is hands-on; AI helps literature and data. | 25% AI helps |
| Consult with engineers or other technical personnel working on construction projects about the effects of soil problems and possible solutions to these problems. Engineer consultations are in person; AI can draft technical memos. | 25% AI helps |
| Develop ways of altering soils to suit different types of plants. Soil amendment trials are field work; AI assists design and analysis. | 25% AI helps |
| Conduct experiments investigating how soil forms, changes, or interacts with land-based ecosystems or living organisms. Soil ecosystem experiments are hands-on; AI aids data and literature. | 25% AI helps |
| Plan or supervise land conservation or reclamation programs for industrial development projects. Reclamation planning is part desk work; site supervision is physical. | 25% AI helps |
| Conduct research into the use of plant species as green fuels or in the production of green fuels. Green fuel plant research is lab and field heavy; AI aids literature. | 25% AI helps |
| Conduct research to determine best methods of planting, spraying, cultivating, harvesting, storing, processing, or transporting horticultural products. Equipment and method trials are field based; AI aids analysis only. | 22% AI helps |
| Conduct experiments to develop new or improved varieties of field crops, focusing on characteristics such as yield, quality, disease resistance, nutritional value, or adaptation to specific soils or climates. Breeding trials are physical; AI aids genomic and yield data analysis. | 20% AI helps |
| Develop improved measurement techniques, soil conservation methods, soil sampling devices, or related technology. Device and sampler design is hardware work; AI helps with software and specs. | 20% AI helps |
| Perform chemical analyses of the microorganism content of soils to determine microbial reactions or chemical mineralogical relationships to plant growth. Microbial lab analysis is physical; AI helps interpret results. | 20% AI helps |
| Conduct experiments regarding causes of bee diseases or factors affecting yields of nectar or pollen. Apiary and bee disease work is hands-on and needs physical presence. | 12% Needs a person |
How soil scientists compare
All occupations (923)
Life, physical and social science (60)
Pay and outlook in the US
- Median pay
- $78,850 a year, May 2025
- People employed
- 15,730 in the US, May 2025
- Projected change in jobs
- +6.7% from 2024 to 2034
Skills to build
- Run field soil sampling and plot trials in person
- Map soil classes and boundaries in ESRI ArcGIS
- Analyze yield trial data in SAS or R
- Earn the Certified Professional Soil Scientist credential
- Draft and verify grower soil recommendations in Microsoft Word
Safer roles nearby
5 related occupations with lower scores.
Questions about soil scientists and AI
Will AI replace soil scientists?
Partly. About 3 in 10 of a soil scientist's working hours go to tasks AI can already do, mostly drafting research papers, slides, and advice for farmers. Field sampling, site-specific judgment, and delivering results to growers stay with the scientist because they need a person on the ground and in the room.
Which soil science tasks is AI taking over?
Communicating research results is the most exposed task, with an exposure score of 50, because AI drafts papers, slides, and course materials. Giving land advice to farmers scores 40, since AI can draft recommendations from soil reports and literature. Literature reviews on sustainability practices score 35 and are increasingly AI-assisted.
Which soil science tasks stay human?
Bee disease experiments score lowest at 12 because apiary work is hands-on and needs physical presence. Breeding field trials and soil conservation device work score 20, since plots, greenhouses, and sampler hardware stay physical, with AI only aiding genomic data and software specs. Microbial lab work also scores 20 because samples must be handled manually.
How can soil scientists stay valuable as AI spreads?
Keep running soil sampling and plot trials yourself, since these tasks score low for AI exposure. Pair that fieldwork with SAS, R, or ArcGIS analysis so you can explain the results to growers. Use AI to draft reports, then check every figure against your field data before it goes out.
Related jobs
Next steps for soil scientists
Check which roles in your team AI can take on
We map your team's tasks, show what AI can do today and build the automations that pay back first.
Related topics
- Jobs AI can't replaceCement masons score lowest, at 8.
- Jobs most at risk from AIProofreaders score highest, at 76.
- High-paying jobs with low AI exposureRadiologists earn a median $420,860 and score 46.
- Entry-level jobs with low AI exposureCement masons need little prior training and score 8.
- Jobs AI speeds up the mostFor natural sciences managers, AI speeds up 96% of task time.
- AI and jobs in Malta15 iGaming roles graded task by task.
- Skills that keep work humanThe skills to build in the jobs AI changes least.
- Jobs A to ZEvery job we publish, by first letter.
- AI exposure by industryAverage scores for 22 occupation groups.
- Write a CV screening software can readHow applicant tracking systems read a CV, and what to change.
- AI for businessesThe tasks to automate first in each team and industry.
- How we score AI exposureThe grading, the weights and every source.