Name: The Sweet Future of Vertical Farming | Hiroki Koga | TED
Uploaded: 2026-03-25T15:49:41.000Z
Duration: 10 min 21 s
Description: Thousands of YouTube videos with English-Chinese subtitles! Now you can learn to understand native speakers, expand your vocabulary, and improve your pronunciation...

Adverbs of degree

So here's a strawberry, but not just any strawberry.

This one can be up to three times sweeter than a regular one, just like the ones I grew up eating in Japan.

As a child, when I came home and I saw strawberries on the table, it meant we were celebrating something special: holidays, birthdays, anniversaries.

Strawberries are considered the king of fruits in my culture, and they're never sold just as strawberries.

They always have their varietal names: Amao, Benihoppe, Toshihotome—any Japanese person can name a few.

Up until now, these strawberries could only be grown under very specific Japanese climate conditions.

But today, you can grow them anywhere around the world using our vertical farming technology.

By the way, does anyone want to try these strawberries?

So today I'm going to try to give it a shot and try to convince everyone here that this technology is no longer some futuristic idea, but

something that is going to mainstream very quickly, much, much sooner than you'd imagine, by sharing some of our secrets.

I came to the U.S. in 2015 to pursue my MBA.

On the second day of my arrival, I went to the local grocery store in California.

I was amazed by the beautiful produce aisle: green, red, yellow, paprikas side by side.

I was especially disappointed with strawberries.

I later learned that this is because the U.S.

agriculture system had optimized everything towards mass production as opposed to quality and flavor.

But this was the moment that I immediately realized that there's a huge opportunity here if I could grow Japanese-quality fruits and vegetables on U.S.

But back then, experts told me that vertical farms were expensive and they could only grow leafy greens: so your lettuces, your kales, your spinaches, because anything beyond that requires bee pollination.

But bees were known not to operate well in a sunless vertical farm environment.

I, we decided to tackle this problem and started building our first farm in 2017 with our own hands, as two non-engineer MBAs.

We nearly lost a few fingers just trying to cut PVC pipes, almost electrocuted ourselves a few times.

But we were really determined to solve one of the world's largest problems that mankind is facing this century: the failing agriculture system.

Extreme weather, lack of water, lack of arable land, heavy use of pesticides, and

just not enough workers to keep up with a growing population.

All of these things have contributed to the global agriculture production cost.

Vertical farms can actually solve all of these problems.

We can recycle most of the water that we use.

In fact, one of our farms used to be a Budweiser factory.

Imagine if you could turn all beer factories around the world into strawberry farms, we'd be much healthier.

But despite all of these benefits, skeptics still say, "Well, vertical farms are too costly of a solution.

It's not a viable business model." I get it.

But think about where computers or mobile phones started.

The real question is, can vertical farms get there, and can we get there quickly enough?

What I can tell you is that a pack of our strawberries that you just ate, they used to go for $50 per pack five years ago.

Now they're $10 available in more than 100 grocery stores on the East Coast.

So you can probably imagine at this pace where we could get to in another five years.

We might even be cheaper than conventional products.

The short answer to that question is that we can innovate so much faster in a vertical farm,

and we can also do things traditional farms simply can't do structurally.

For example, in a traditional farm, you can only experiment once a year during the season.

At Oishi, we have dozens of grow rooms where we can control every aspect of the environment, from things like temperature, humidity, CO2 levels, all the way down to even light spectrums and wind speed.

So we can conduct experiments so much more efficiently in a controlled manner.

And on top of that, we can start these experiments anytime during the year because we're not impacted by the outdoor seasonality at all.

And using this approach, we were able to go from just a few berries on a plant to five times of that in just a matter of five years.

If you try to accomplish something like this in a traditional farm where you can only experiment once a year, it could have taken us nearly 500 years to get to the same outcome.

And using this hyperspeed experiments, we were also able to accomplish three major breakthroughs: bee pollination, AI, and

I know this is everyone's favorite topic.

You need bees to grow most produce: strawberries, tomatoes, melons, peppers, you name it, because the bees will pollinate the flowers that then become the product.

But bees, they need the sun and the natural environment to navigate.

So in a vertical farm, they wouldn't even come out of the hive.

Even if they did, they will drunk fly and can't find the flowers, just like how drunk humans are completely useless.

On top of that, you need somewhere between six to 15 visits by a bee to fully pollinate a strawberry flower.

Anything less or more will result in mispollination, meaning you won't have product.

So for these reasons, an average Japanese farm, a traditional farm, is set to have a pollination success rate of somewhere between 60 to 70 percent, meaning three to four flowers don't even become products.

In order to solve all of these bee problems, Brendan and I, we interviewed many, many apeologists, experts.

We spent so much time in the farm observing the bees ourselves, getting stung at times.

Or ultimately, we mapped out every environmental parameter of outdoor farms versus our indoor farms and went one by one.

And after two years, finally figured out the bee recipe.

So today we have bees, sober bees, in our farms doing all the pollination for us.

And what's more, with the help of data science and AI, our pollination success rate is above 95 percent, meaning more than nine flowers out of ten become berries.

So we're much more efficient than conventional methods.

We have these self-driving camera carts that's taking real-time visual data of every single plant in the farm, which helps us not only to determine yield and plant health information,

but also how much bee activity is needed that day to help with the bee pollination.

These AI-powered robots can automatically detect only the ripest berries, pick them automatically, and they can work 24/7.

Subtitles ListPlay Video

The Sweet Future of Vertical Farming | Hiroki Koga | TED

ultimately

massive

figure

grocery