AI in vertical farming could transform global agriculture | The Excerpt
On a special episode (first released on November 20, 2024) of The Excerpt podcast: AI applications in vertical farming have the potential to usher in a new model that not only yields a high volume of quality product, but also offers a much smaller carbon footprint in terms of resources and the space necessary to deploy them. By incorporating AI, robotics and vertically stacked crops in controlled environmental conditions, the vertical farming model might be a gamechanger in the agricultural sector. Hiroki Koga, co-founder and CEO of Oishii, the company behind the world's largest indoor vertical farm, joins The Excerpt to share his insights into how AI is transforming farming potential.
Hit play on the player below to hear the podcast and follow along with the transcript beneath it. This transcript was automatically generated, and then edited for clarity in its current form. There may be some differences between the audio and the text.
Podcasts: True crime, in-depth interviews and more Paste BN podcasts right here
Dana Taylor:
Hello and welcome to The Excerpt. I'm Dana Taylor. Today is Wednesday, November 20th, 2024 and this is a special episode of The Excerpt.
Artificial intelligence has had a revolutionary impact across numerous industries that has been felt across the world, but one industry in particular that could soon beginning an AI-enhanced makeover is global agriculture. AI applications have the potential to usher in a new farming model that not only yields a higher volume of quality product, but also offers a much smaller carbon footprint in terms of resources and the space necessary to deploy them. Vertical farming may provide such a solution. By incorporating AI, robotics and vertically stacked crops and controlled environmental conditions, the vertical farming model might be a game changer in the agriculture sector. Here to help us unpack this and to discuss the potential impacts on farming around the globe, I'm now joined by Hiroki Koga, co-founder and CEO of Oishii, the company behind the world's largest indoor vertical farm. Thanks for coming on The Excerpt, Hiroki.
Hiroki Koga:
Thank you for having us today, Dana.
Dana Taylor:
Let's start with the basics. What is vertical farming and where did this idea first originate?
Hiroki Koga:
So vertical farming is a new way of growing the crops indoors without relying on any external sunlight. So a lot of the times we're inside a warehouse-type building, and basically, the idea is that we can grow crops anywhere around the world as long as there's access to electricity without being impacted by the outdoor seasonality.
This technology itself was first commercialized in Japan probably more than 15, 20 years ago. And so at one point we had probably three to 400 vertical farms commercially operated in Japan where I'm originally from. But the industry kind of failed because these farms could only grow lettuces and leafy greens, which were a commodity. Back then, the cost of production of vertical farms were very expensive, so they couldn't compete against the outdoor.
Dana Taylor:
I know that this wasn't your background, farming. What drove you to pursue this technology and to start your company, Oishii?
Hiroki Koga:
I joined this industry as a consultant, and I've looked at many of those facilities and a lot of them failing. But then in 2015, I came to the US to pursue my master's here in the US, and that's right around when vertical farming industry started to take off here in the western world, primarily driven by the sustainability pressure onto this agriculture industry. And vertical farming was thought to be one of the potential solutions as we require significantly less resources compared to conventional farming.
Dana Taylor:
So what drove you to pursue this technology and to start your company Oishii?
Hiroki Koga:
Because I had a lot of experience in this industry in Japan, where we've been doing this 10 years before the United States, I had a lot of experience and at the same time, I realized that there's a huge opportunity here in the United States as the quality of the crops that I experienced in the United States were not, to be blunt, as fresh and as tasty as what I was used to eating in Japan. So I thought if I could grow Japanese quality produce, especially things like strawberries or tomatoes where I saw the biggest quality gap, and I thought it would benefit both the consumers here in the US but also as a Japanese, I can share the culture that I grew up with with the rest of the world.
Dana Taylor:
What are the various ways that your company is integrating AI into your vertical farm?
Hiroki Koga:
We use AI on a lot of different types of technologies. For example, unlike a traditional farm, because we are setup in an industrial setup, we're able to monitor visual data of every single plant in our farms. And also, we can also monitor environmental data like things from temperature, humidity, CO2 levels, all the way to wind speed. So we have this huge, massive, massive data asset, which we will run against our AI algorithms that will allow us to analyze what the health condition of the farm is every day. And based on that data, we would either use that data to communicate with our bees who's in our farm doing all the pollination for us to really optimize the pollination condition for those bees using our AI driven data, as well as optimize our growing environment for the plants based on all the data that we acquire on a day-to-day basis.
Dana Taylor:
An ongoing concern within the agriculture sector is sustainability. In what ways does the vertical farming model offer a sustainable path forward as the industry grows?
Hiroki Koga:
Yeah, so when you think about what goes into current agriculture, you need stable weather, land, you need access to water, you need access to labor, and then also use of pesticides. All of these five things were a prerequisite and they were relatively easy to source. But as you can imagine, none of these are easy to source anymore, and as the population is exploding, we just won't have enough land or enough water to keep up with a growing population.
And vertical farming, as you can imagine, one, we don't even need farmland. Our current facility used to be a plastic factory that got refurbished into a strawberry farm. We can recycle more than 90% of the water that we use. We don't use any pesticides. We can automate a lot of the processes because we are like a factory set up. We use significantly less resources. The only thing that we consume more than conventional farming is electricity, but with our most recent facility, we have a huge solar power generation right next to our farm. So we're powered by clean energy as well there.
Dana Taylor:
With the various measures of controlling the environment that virtual farming allows, what sort of advantages does that offer in terms of crop yield?
Hiroki Koga:
The toughest part of agriculture these days is getting the most out of the seeds that you have, and a lot of the times it's the environment that prevents us from maximizing our yield. Sometimes you would have extreme weathers like storms or rain, drought, and these are things that you don't have to worry about in a vertical farm. So as long as you know the perfect growing recipe or the growing environment for each type of crops, then all you have to do is recreate that perfect environment every single day in a vertical farm, which is so much easier to do than in an outdoor farm. And that's what allows us to really maximize the yield. And not only yield, but also the quality potential of each genetics that these crops have.
Dana Taylor:
We've talked about how AI can be applied within the virtual farming infrastructure, but what about the robotics that are doing the manual labor in this equation? How many different robots are employed within a vertical farm? What are the possibilities or what are the responsibilities here?
Hiroki Koga:
If you can imagine what a car manufacturing factory looks like today versus 100 years ago when it was all assembled by human hands, that's kind of the similar process and evolution that we're going through right now with agriculture. So all the way from seeding the plants to growing them, controlling the environment automatically, picking the berries, putting them in packages, moving the packages to the packaging room, assessing the quality and then putting onto the trucks. All of this should be automated in the coming years.
For us at this point, we already have a lot of those procedures automated, including the harvesting component. So we have AI-powered robots that can automatically detect which berries are ripe and ready to go. We still have humans in the farm as well, but a lot of our berries are actually already picked by robots at its perfect ripeness, and so they're not even touched by a human hand all the way until they're consumed.
Dana Taylor:
When will average consumers see fruits and vegetables grown on vertical farms at their local grocery store, and what will that cost look like off the shelf?
Hiroki Koga:
I would say their cost is probably somewhere between 30% to 50%, sometimes 100% more expensive than conventional products. But I think the benefit that we get is it's usually fresher and the quality is very consistent because most of the time they're harvested locally as opposed to grown in California, shipped all the way to the East Coast where I live. We have strawberries, and because our strawberries are not just any strawberries but very sweet and much higher quality, our cost is a little bit higher, but a pack of our berry now goes for approximately $10. So they're not the cheapest strawberries, but they're flying off the shelves and I think people are seeing the benefit of our product.
Dana Taylor:
Hiroki, is there a way that vertical farming can help alleviate the global health crisis? How do you see this as being a viable solution on such a large scale?
Hiroki Koga:
Yes, I certainly believe that, and the reason is right now the agriculture supply chain system is flawed. And for example, in the United States, 90% of the strawberries are grown in California and they're shipped across the country to places like New York. And a lot of the times these strawberries are traveling for more than a week. And with vertical farming, we will be able to grow crops anywhere around the world right next to where consumers live. And yes, they are not the cheapest product today, but as you can imagine how computers went from $100,000 20, 50 years ago to now something that's affordable for everyone, just a matter of time that vertical farming will actually become mainstream. Especially as our cost is going down very quickly and cost of conventional farming is only to go up from here. So it's evident that we will be cheaper and we will also have fresher and higher nutrient products that will be accessible to everyone around the world.
Dana Taylor:
What can you tell us about how vertical farming has the potential to increase pollination among bees and boost the bee population? I know you have some IP here that of course we respect your right to not divulge, but what can you share with us?
Hiroki Koga:
I think the biggest limitation in vertical farming up until we came in was that people couldn't grow anything beyond leafy greens because most crops beyond leafy greens require bee pollination, but bees were known not to operate well in a [inaudible 00:10:44] vertical farm environment. And so we spent the last seven years really trying to understand the bee physiology, keeping them happy in our farms and figure out how to make them believe that they're in nature. And with this, we should be able to grow pretty much any types of crops.
And in a vertical farm, because we can collect data so much more precisely compared to outdoor farms, we're able to collect the health data of every single plant. We're able to collect the health data of the bees and reconcile those, and on a day-to-day basis readjust the amount of activities that happens in our farms. So in a traditional farm, bee pollination success rate is set to be somewhere around 60 to 70% when it comes to, let's say, Japanese strawberries. But in our farm, our bee pollination success rate is above 95% thanks to AI and all the robotics that we have.
Dana Taylor:
Then finally, what's next for the future of AI in farming?
Hiroki Koga:
The speed of and the evolution of AI has been truly surprising to many people just in the last couple of years alone. We've been able to accomplish pollination success rate going from 60% to 95%, which means our costs went down by almost 30% just on that pollination part alone. As AI develops and we can integrate AI into a lot of different processes in addition to just bee pollination, but automation of picking, packing, quality assessment, all of these things, I'm sure that our costs will go down dramatically, and it's definitely going to accelerate or help us get to solving that cost problem much quicker.
Dana Taylor:
Hiroki, thank you so much for being on The Excerpt.
Hiroki Koga:
Thank you for having me today, Dana.
Dana Taylor:
Thanks to our senior producers Shannon Rae Green and Kaely Monahan for their production assistance. Our executive is Laura Beatty. Let us know what you think of this episode by sending a note to podcasts@USAToday.com. Thanks for listening. I'm Dana Taylor. Taylor Wilson will be back tomorrow morning with another episode of The Excerpt.