Free sampler. Re-scored every six months.
Farm management scores 4.0 out of 10 for AI exposure, where 10 is most at risk. Horticulture and landscaping also score 4.0, conservation and field ecology 4.7. The most exposed track is precision agriculture at 6.2 — the technology-forward version of the field, not the traditional one. This is the category where the biggest genuine unknown sits.
The short answer for parents: reasonably protected, and the least certain scores in the index. Living things do not behave like production lines, which is the protection. But agricultural technology is advancing quickly in exactly the controlled settings where automation always arrives first, and this is where our long-horizon caveat about physical AI has the most bite.
Every career in this index is scored 1–10, where 10 is most exposed to AI. Same six factors, same weights, applied identically to a farm manager and a paralegal.
| Agriculture track | 2023 | 2025 | Now | 3-yr move | Band |
|---|---|---|---|---|---|
| Farm management / mixed farming | 3.3 | 3.7 | 4.0 | +0.7 | Low–Mod |
| Horticulture / landscaping | 3.5 | 3.8 | 4.0 | +0.5 | Low–Mod |
| Conservation / field ecology | 4.0 | 4.4 | 4.7 | +0.7 | Moderate |
| Environmental consultancy | 4.7 | 5.0 | 5.2 | +0.5 | Moderate |
| Controlled environment agriculture | 5.2 | 5.5 | 5.8 | +0.6 | Moderate |
| Precision agriculture / agtech | 5.6 | 5.9 | 6.2 | +0.6 | Mod–High |
2023 and 2025 figures are reconstructed using current methodology, not archived from past editions.
For scalethe median career in this edition scores around 5.5. A service electrician scores 2.5. Veterinary practice scores 2.7. A business analyst scores 7.6.
Note the ordering. The traditional, outdoor, hands-in-the-soil versions score best. The technology-forward and indoor versions score worst. That is not a coincidence, and it is the finding this profile is really about.
Not in the field. It is transforming how field work is done, which is a different thing.
Agriculture is one of the few fields where the technology is largely being adopted by the people doing the work rather than imposed on them — which changes the shape of the risk considerably.
How farm management rates against each. Ratings are 0–10 on each factor's own terms.
So you can see what the analysis actually looks like.
How much of this job can AI already do? — rated 6.0
In most professions this factor measures a threat: how much of the work could be taken away. Agriculture is the clearest case in this index where the same rating means something different, and it is worth being precise about why.
A 6.0 here reflects real capability. Yield modeling, variable-rate input application, satellite imagery analysis, irrigation scheduling and compliance record-keeping are all genuinely automated, and the technology is good.
But look at who is buying it. Precision agriculture tools are sold to farmers, adopted by farmers, and operated by farmers. The technology arrives as equipment — a smarter sprayer, a better sensor, an imagery subscription — rather than as a replacement for the person deciding what to plant and when to act.
This is structurally different from, say, a paralegal, where the automation is bought by the employer and reduces the number of paralegals needed. In farming the automation is bought by the operator and increases what one operator can manage.
That distinction shows up directly in the score table, but not in the direction people expect. Precision agriculture rates 8.0 on automatability and scores 6.2 — the most exposed track here. Because the more a role is defined by operating the technology rather than by physical presence and judgment on the land, the more it becomes a data job that happens to be about crops. And data jobs score badly in this index.
And controlled environment agriculture — glasshouse, vertical, indoor — scores 5.8 for the reason that runs through this whole index: it is farming in a predictable environment, which is where automation always arrives first. It is the agricultural equivalent of the production line versus the service call.
The general lesson: ask whether automation in a field is sold to the worker or to the employer. When a person buys a tool that makes them more capable, the tool tends to raise their value. When an employer buys a tool that replaces a function, it tends to lower it. Same technology, opposite effect on the person, and the score alone will not tell you which is happening.
Ranked by exposure, safest first:
It is a genuinely durable pull for a particular kind of person, and it is poorly served by careers advice.
The environmental and land-management side is growing for reasons entirely independent of AI — climate adaptation, regulation, restoration funding and land-use change. That demand is real and not cyclical in the way construction is.
Farming itself is capital-intensive, and the honest obstacle for most people entering it is access to land rather than automation. Farm management roles, contracting and agtech provide routes in that do not require inheriting or buying a farm.
This is also the category with our largest genuine unknown. Physical protection is the strongest shield today and the one we rank least durable, and agriculture sits in the middle of that question. Field robotics is advancing fastest in orchards, glasshouses and row crops — structured settings — and slowest in mixed farming, livestock and uneven terrain. A student entering now will work for forty years, and the boundary between those two categories is likely to move.
This sampler tells you where agriculture and environment stand. The full profile tells you what to do about it.
Most families are weighing two or three careers seriously, and a few more they haven’t ruled out. Pick the ones you need.
Each includes a short version written directly to the student and the technical scoring appendix. Spring 2027 updates of whatever you buy are included.
Read one complete profile free → We publish computer science in full so you can judge the depth before buying anything.