Is AI the future of farming?

Ever since the end of the Stone Age, humans have created farms to feed themselves and support large population centers. Over the years the methods of farming have largely remained the same, but with new machines to do things many times faster than people can. For the first time ever, it’s actually humans that could be replaced on farms instead of just getting an upgrade on the tools we use, and sometime in the future it would only take a handful of people to farm hundreds of acres of land.

What is AI and how would it affect the farming world?

AI stands for artificial intelligence and is just an advanced computer program that has the ability to think and work out problems for itself. The usual process is to present the program with a load of information and give it the ability to process that information and come out with an answer to something, a process that can be done almost instantly. If the program was given access to the Internet and also thousands of journals and papers about farming, it would be able to remember everything it was exposed to and be able to compare all the case studies and information it read to give the best answer when asked a question.

But how would this affect farming?

Even though we have a huge amount of advanced machinery and pieces of equipment that can do the jobs of hundreds of people, it is still the farmers themselves who choose when to plow the fields and harvest the crops. People are needed to drive the tractors and decide when to apply fertilizer or pesticides, but all of these decisions are down to human judgment. We already have the ability to build remote control and self-driving vehicles, and it won’t take long until someone applies this to a tractor and combine harvester.

If the AI program was advanced enough, it would be able to do everything at exactly the right time with pinpoint accuracy. The fields would be mapped out to the millimeter and the drone tractors could just be left on their own and would be able to run nonstop, or until they needed recharging at least. A single room of operators could control dozens of tractors across several farms and simply create a list of protocols for them to follow and then just leave them to it. The job losses within the farming sector would be huge, but this method of farming would also come with its own list of benefits.

Are we close to AI-controlled farms?

No, no we are not. realistically it would take well over a decade before drones even came close to starting to replace people. The current leap forward in AI technology is impressive but there’s still a long way to go before we can use them to effectively produce food. All of the ideas around AI farming like automated tractors and flying drones that can identify and sap weeds with a laser are all in their infancy and years away from being produced.

As with all new technologies, upon their initial release, the cost to acquire a farm’s worth of equipment will be out of the reach of anyone apart from the government. There is a popular misconception that farmers earn a high wage, with all that land they own and tons of crops being produced must mean a high return, but in reality, this is far from the truth. The average wage of a farmer in the UK at the start of 2024 was just under £30,000 a year, which is a long way from being able to afford the latest self-driving drone tractor technology.

There are many factors why people will never be completely replaced by AI on farms because we have no option but to keep farms running, and if people can’t afford the new technologies then they’ll carry on the same way they always have. The government won’t be able to force people to upgrade and will eventually start to offer schemes where farmers can get hold of AI farm equipment for cheap, but it will take many years for the technology to be perfected and come down in price enough for them to be seen on regular farms.

How would AI be used in farming?

The higher end of the development scale would be self-driving tractors and autonomous flying drones that could control pests and weeds and monitor the fields, but this level would be the last to be available. Initially, AI would be used in a simpler sense, such as AI-assisted monitors that could give directives and prompts to the driver, or point out the first signs of crop disease or an infestation.

A flying drone with a camera would be able to quickly scan the whole field and report information back to its operator. Basic assistance in the form of information would be the initial helping hand that AI offers, but this alone comes with a few problems.

Computer literacy

This will be one of the biggest initial challenges to distributing AI within the farming community. Anyone born within at least the last decade would have been heavily exposed to computer technology and be familiar with how to use various types of user interfaces, but there are those who just can’t get on with computers. Some people simply don’t like using them or do not need them and would find it very difficult to go from very basic levels of technology to calibrating an autonomous tractor to drive around several awkwardly laid out farm fields. The initial lack of understanding will stop or slow down many people from making the leap over to AI farming, and that’s even if they can afford it.

How is AI currently being used for farming?

At its current level, AI is mostly being used in the prototype stage for land monitoring. There is a company that has produced a trapping device to attract insects and rodents using pheromones and various baits, and after they have been caught, a camera will take a quick photo before letting them go and resetting the trap. The AI program will identify the creature and log information about it, such as which areas it appears in and how often. All of the traps send information to a central program that can then inform the farmer about which pests are in which areas and if any of them are likely to cause problems.

Another interesting creation is soil monitoring poles which can detect the areas of the field that hold water and the areas that need treatment. After it rains, the poles detect how far down the water reaches and they send that information to a program that can map out the field’s water tables. This may not sound like much but could save millions of gallons of water each year by watering certain areas less, and more advanced poles are being developed that can detect nutrient and soil quality levels, telling the farmer exactly when to perform some kind of treatment.

When it comes to weeds growing on a farm field, the overall reduction in crop yield can be as much as 50%. Current treatment involves spraying the crops with chemicals that are designed to only affect the weeds, but this is expensive and often requires buying various types of weedkillers. There is currently a laser-based weed-killing machine that can be towed by a tractor and is supposed to destroy 99% of weeds at a rate of 2 acres per hour. The advantages of this are huge, and not just in the time saved or the increase in crop yield, but it can run on nothing more than electricity and could greatly reduce the production of chemical-based weed killers.

The problems with using AI on farms

Electricity distribution
Some of the larger farms use multiple drivable machines and something the size of a combine harvester would presumably require a large amount of electricity to operate. All of these machines would need to rapidly charge, presumably all at the same time at night, leading to a massive demand for power for every single farm that used them.

Current power storage and creation
We are almost at our limits when it comes to how much power we can store in an average battery. If we want to create higher levels of power storage then a new type of battery needs to be invented because the current lithium-based ones are simply not high-capacity enough. The UK and most other countries that are using electric cars have found there to be major problems when it comes to charger distribution and battery recharge times, and having a load of very high-drain vehicles in concentrated locations all around the country won’t be easy to accommodate.