▲ 76 points
back
113 comments
> When human bodies are scarce, as they often are in rural Australia, machines are created to fill the void.
I think they meant where labour costs are more than their ability to pay
Automation is seen in the more advanced countries, ones which lack an oppressed underclass like the US. There's much more automation on farms. Robotic milking is widely used in Australia, but less popular in the US. [1] New Zealand has the best automated slaughterhouse technology.[2] Those jobs are mostly done by illegal immigrants in the US, and the companies hiring them are screaming about stricter ICE enforcement.[3][4]
[1] https://dairy-cattle.extension.org/dairy-robotic-milking-sys...
[2] https://www.youtube.com/watch?v=mdq1uBCI2Kw
[3] https://www.usatoday.com/story/news/2025/06/23/dairy-industr...
[4] https://tucson.com/news/national/article_fe9b9724-1526-557b-...
> and the companies hiring them are screaming about stricter ICE enforcement.[3][4]
Your statement matches nothing listed in those linked articles. [3] Talks about how the dairy industry is concerned as it relies on immigrant labor. [4] only quotes an "official of the plant" who says they are against illegal immigration, a personal opinion.
[4] also amusingly paints the immigrants as tech savvy identity thieves which is laughable. Likely, it is whatever shady entity the employer uses to hire illegals allowing them to deny involvement.
This is secondhand and not firsthand knowledge, but one thing I've heard from a number of agtech founders (and sometimes farmers when we do diligence calls with them) is that hiring employees is very hard because of seasonality. E.g. you might need 100 workers most of the year, but 400 for a few weeks when it's picking season. And of course if you are in an area known for a single crop, all of the other local farms have the same seasonal labor need.
For these types of businesses, paying less for labor is a secondary goal and just having enough labor (human or machine) is the primary goal.
The US sugar beet harvest is like that, but seems to have structured itself around that.
Lots of websites recruiting people with RVs to come by and work the 2-3 week season. Others stay in motels or live close enough to drive.
https://www.google.com/search?q=beet+harvest+working
https://www.theunbeetableexperience.com/sugar-beet-harvest-f...
Some places have/had the schools close for the harvest season (and in others, for the start of hunting season).
Rural Australia also just has a desirability problem. Rural towns regularly offer significantly above market rate for GPs and teachers, but rarely ever fill the posts.
Sounds like they aren't actually market rate if the labor market is rejecting the prices at those levels.
Agriculture is bottom of the barrel business because no political power worth its weight allows price gouging. Food has to be kept as cheap as possible because otherwise the economy doesn't work. The workaround for this is subsidies, but those don't scale. You can't agriculture your way into Google Ads, the money printing machine.
Could you elaborate on "subsidies don't scale"? In the US, farm subsidies are a huge chunk of our budget and, to the best of my understanding, help keep food prices low. I'm not informed enough to know if it's an inefficient solution though.
You can't anything your way to miracle profits without destroying society in the process.
I suppose destroying society for profit is top of the barrel for people like you.
Would I be correct in assuming that your name comes from 'The Point'? What a classic!
I also appreciate your point about the reality of agriculture, I think too often people miss just how narrow the margins there are and why farm life has always been so financially unstable. It used to be the weather that could make or break you, and it still is, but not world commodity markets, the price of fuel and fertilizer, and trends in a number of areas can all do it as well. Add in the top-down pressure to keep prices low and bottom-up pressure to target migrants and... what a rough mix.
Human bodies are scarce because you can't pay most people to live there. They're not mutually exclusive.
Yes but the point of the poster is the framing of the conversation.
There is no labour shortage, there is a salary shortage. And we, as the collective global community of serfs and plebes must realize this, and call it for what it is.
There is currently a massive teacher shortage in Ireland, with something like >1800 unfilled posts and we're approaching the new school term. There is no teacher shortage, there is however an abundance of catholic-church-controlled schools with overly restrictive hiring policy, with many newly qualified teachers not really interested in becoming involved in religious things. Teachers are paid well here, but clearly not well enough for many of them to be willing to subject themselves to draconic requirements such as needing to provide catholic teachings to kids taking fucking math.
I dream that AI robotic drones can reduce our dependence on vulnerable monocultures. Rather than a monoculture processed by million dollar combine harvesters using bulk techniques, a polyculture harvested by a swarm of small cheaper drones that work at the individual plant level.
I think this isn't going to hapen. The economy of the scale favors monoculture and it combines well with automation. The most likely outcome will be that combine harvester not requiring a driver and a synchronous operation of multiple harvesters as a robot swarm such that much bigger fields than currently possible could be worked: monoculture at a much bigger scale.
Polyculture has two very significant advantages.
- Because you're pulling multiple crops off a field instead of one, the value of those crops is significantly higher.
- The crop is much more resilient. The different crops will have different tolerances to drought, disease, insects, wind and other disasters that commonly affect farming.
When farmland was cheap you could increase your profits per employee by buying more land and bigger machinery. But these days farmland is expensive and farmers are concentrating more on increasing yields per acre. In my area farmers spend > 10x as much per acre on inputs like fertilizers and pesticides than they used to in the 90's.
Polyculture will be hard and expensive, but if it lets you double profit per acre, farmers will do it.
> combine harvester not requiring a driver
Already pretty much the case. They may have a driver as a monitor/troubleshooter but for planting and harvesting it's all optimized down to the inch with GPS guidance steering the machinery.
Two of my friends who farm already have self driving planters and harvesters, controlled by GPS, that lay down different kinds of seeds and fertilizers depending on location within their fields. The farmer is there to fix things.
> The economy of the scale favors monoculture
I mean it does, until it doesn't and you loose all your crops/soil/water.
Monocultures degrade the soil, and you need to put more and more input in to make sure grow. Because there is no ground cover between crops, wind takes away your soil. If you mess up/mistime your pesticide/herbicide input, you'll loose all your crop. The more you grow, the more the soil turns to sand, requiring more fertiliser input, which mean more costs.
If you take the yield per m2 of an allotment, and compare it to a really high quality agricultural field (ie Lincolnshire fens, or some prime land in the USA.) the allotment will have a much higher yield with less input. I don't have the figures on hand, but you'll need to turn to the bearded hippies at the henry doubleday to get precise numbers.
If, through automation, you could have three or four crops in the same field, providing cover all year round. you could then start to keep your soil, because the wind cant get at it (crop dependant). you can grow sacrificial crops for pest control, meaning less pesticide input. This means more worm, which will stop flash flooding (more water permeability without ploughing)
But all that requires high speed "pick and place" weeding/harvesting machines. Those are going to be complex and expensive until scale kicks in (think how complex tractors were, or threshing machines)
There is a solid equilibrium in a commercial monoculture and artisanal and subsidised polycultures for insurance. The transition will never be easy. But at least the groundwork is there.
The future we’re already in should be characterized as cheap complexity. As an example, in manufacturing one offs were cost prohibitive, now it’s close enough to mass manufacturing cost that in many cases there is no point taking that extra step. Also people are increasingly concerned with food provenance, I would love to have a local automated greenhouse growing hairloom vegetables.
I hope it goes that way bit I expect the opposite. They way big ag doesn't show much respect for nature I expect them to go even more into monoculture and using robots to wipe out all insects and plants they don't like.
They'll do it if it's profitable. More output per acre is a pretty big motivation.
Grad students will work on the problem for the ecological benefits. Then big ag will scoop it up for the profit motive.
Or at least reduce the need for pesticides
> what farmers want: technologies that may not have a lot of bells and whistles, but can reliably take a task off their hands.
I'm sure they are not alone.
>> but can reliably take a task off their hands
Tasks can reliably take off farmers' hands.
Every year our state fair has dioramas to demonstrate the risks of mangling, amputation, and death.
I don't really want to see hallucinating LLMs making this even more dangerous despite how good it may make VCs feel. It's possible for it to not be a closed system but have someone's gain mean someone else's loss. (Arguments over closed vs open system are a sophistic tactic to ruin other people's lives for money.)
TIL I am a bugs farmer, not a software engineer.
I am a backlog, hear me score.
I don't need (more) automation on this farm, I need (lots) of help shifting product. I can make a lot more than I'm able to sell!
> Our research team conducted more than 35 interviews with farmers, specifically livestock producers
Not agriculture, cattle
And the cattle farmers massively benefit from these things. There are now custom monitors being put on to cattle that constantly measure how much they eat, how much they walk, body temp, etc. If they fall out of expected range the system will call a vet automatically to come treat them as they're obviously sick. This is keeping the cattle healthier and the production numbers high.
Farmers are also really getting into things like drones for pesticide application as it's faster, more accurate, and has less waste.
It's happening whether or not people want to believe it, from Texas up to Illinois/Wisconsin and everywhere inbetween.
hey, if you want to get away from big tech and you're not scared of rural life -- farmers need IT folks. They have racks of Supermicro servers on premises that need maintaining because they can't trust the cloud // internet can be unreliable. They have sensors and repeaters on the field perimeters that need maintenance.
There's a lot of IT work to be done in rural America with a slower pace of life if you're not afraid of coexisting with country folk. For those of us that grew up there and migrated to the city for corporate IT jobs -- I expect we'll be going back to our roots and serving our communities with these skills instead so we don't have to deal with AI screened jobs and take-home programming tests
The way they are treating cattle as pure machines to optimize production I can only hope synthetic meat will come soon.
"Cattle production is the most important U.S. agricultural industry" [1]
"Animal husbandry, is the branch of agriculture concerned with animals that are raised for meat, fibre, milk, or other products" [2]
[1] https://www.ers.usda.gov/topics/animal-products/cattle-beef/...
A large industry, but also a much smaller group than ‘live stock producers’.
I worked on strawberry picking robots for a few years as a field robotics engineer. I don't know about Australia but for American fruit growers and to a lesser extent, British ones, the tasks that are actually economically valuable are shockingly hard technically.
Picking strawberries at a rate that beats human pickers, given the endless amount of variables, is insanely hard. Why not pick another task like data collection, you ask, because its just not important enough. So what you have is a bunch of techbros making SISP's and attracting VC's who are hopped up on SaaS sugar rushes, the tech doesn't work and the company folds, and the farmers get disillusioned and less amicable to the next entreprenuer.
In 2025, we're in like the third cycle of this.
Only the third? I remember reading an article about a plant orientation sensor for an automated farm robot in 1993
Yeah maybe more. I directly participated in the 2019-2023 hype cycle for AgTech
In developing countries, farming is left for those who has no opportunity or ability to do any other better work. Often uneducated, poorer, living in remote places without basic facilities.
Result? The farmers are forced to do extensive use of chemicals in farming, low nutrition, high yield GM/hybrid varieties and adulteration of whole foods. The food that comes out is laced with chemicals, poor quality varieties. And the urban consumer pays for their negligence of the farmer.
In low income economies over 50% of population 10 years and older own a mobile phone.
https://www.itu.int/itu-d/reports/statistics/2024/11/10/ff24...
Access to information even for the global poor is almost universal. But like the old joke about the ag extension agent trying to get a farmer to improve his practices with information, “Heck, I don’t farm as well as I know how to today.”
Who is going to sell me a system where I stick a couple of Zigbee sensors in my home vegetable garden and I take pictures of the garden once or twice a day and the system uses this info plus weather data to tell me what I need to do the next morning ?
Funny talking about the Suzuki Sierra, the Toyota Landcruiser 70 series is exactly this. It's a $90k (AUD) ute that has almost no bells and whistles but its basically bulletproof.
Waiting for the title to change to "The future of farming"
I wonder how long GMO (the in the lab kind) took to get adopted after surely initial resistance.
Will we see AI-free food? Will it become a part of being “organic”?
GMO was already adopted before the resistance even manifested. It was a little bit more crude in the early days. Irradiate a field and select desirable traits among the mutants.
Something like AI hallucinated produce? Random taste tomatoes anyone? Leaving the joke aside, probably GMO will probably accelerate with the help of AI and will still be called GMO foods.
"Human made" will be a sign of quality, like the "organic" label.
Agriculture is one of the few fields where "shit in" is a expected, good thing. A fertiliser on the fields.