We spend 1.6 billion tons of food every year while 25 million are hungry and one billion are malnourished.

Can AI fix it?

If a startup in Berlin succeeds, it’s possible.

The global food supply chain is very confusing. Tens of millions of farms feed millions of shops and restaurants, which in turn supply nearly eight billion people their daily food. Plus of course there are transportation companies, wholesalers, distributors, processors, and shipping companies. Put it all together, and you have a large network of producers and consumers who are joined by everyone in between who literally covers every human being on the planet.

And they all, of course, deal with perishable products.

The complexity and chaos is the startup company SPRK.global, is trying to fix it. The company is one of them eight winners of 2,400 participants in “the world’s largest beginner competition for entrepreneurs overcoming global challenges,” Extreme Tech Challenge.

The goal: use AI to understand food flow and reduce waste. That should cause less hunger and overproduction.

“Half of the food produced will be wasted sooner or later,” CEO Alexander Piutti told me recently on the TechFirst podcast. “Once you start to understand patterns – why there are cases of food waste – you understand these patterns and see them appear regularly … we can switch from reactive to proactive, anticipate, to predict with certain probabilities.”

Food waste is an ecological problem and also a human problem. Overproduction uses resources such as fuel, water, fertilizer, increasing greenhouse gas emissions in the process. And when food is wasted, you need landfill space and there is more CO2 production.

“Food waste is one of the biggest emitters of CO2,” Piutti said.

The question is: can AI solve it?

But before AI enters, the system needs to understand the supply chain and the economy. For example, if there is an oversupply in one area, that does not mean that the store or distributor with too much wants to give it to competitors with too little. On the other hand, they tend to give it to food banks and other NGOs that can give it to those who need it and will not be able to afford it. So SPRK needs rules like this to function in the real world from actual economic conflicts.

“Once we have these rules, we can inject them into technology,” Piutti said. “This technology is taking over … and matching between oversupply and demand … getting smarter over time.”

In a way it’s similar to Google’s Loon Project, where Self-regulated internet balloons provide cellular connectivity and web access in rural Kenya. Where they drift in different winds set at different heights, and machine learning algorithms that deploy balloons have learned over time – and are still learning – how to navigate to maximize the range of territory needed.

One place to start is the food bank.

They usually operate with telephones and perhaps spreadsheets, but software attacks as a service that takes over so many industries, basically, software that operates businesses has not touched food banks or similar NGOs. So SPRK is building software that they can use to manage their own operations and collaborate: one has too much, the other is too little. Sharing will now be replied in the future.

In addition, said Piutti, the software will give them a better way to access food at lower prices.

“They buy food in a very normal way, they don’t get a discount,” he told me. “If we can connect the dots conceptually and say like, what if we distribute this excess supply of food to people who need it … they become volume partners.”

In other words, most NGOs who distribute food buy it on the open market. If the distributor has too much, SPRK can access it – with the condition that it will not re-enter the commercial market – and give it to food banks and others at a much lower price. That saves NGOs around 50%, reduces food waste, and provides operational income for SPRK.

“So, win, win, win the situation,” said Piutti, who worked for Yahoo at the time and gave advice to many other startups. “This is how we think of solutions, you know, without generating channel conflicts, to be transparent to our supplier partners, because they have to agree.”

The goal is not to reinvent the wheel: just to use what is a little smarter.

While SPRK began in Berlin, the company has global aspirations from the US to Asia, which could eventually become a large cloud-based platform for fair, efficient and effective food distribution. At least, excess food.

“It’s like a big platform, just call it ‘Amazon of Food Oversupply’ if you like,” Piutti said.

This is a big vision and there are many things that must be built before it’s reality. Over time, if SPRK succeeds, it is likely to surpass distributors and NGOs to producers and consumers, inject intelligence and predictions into agriculture and predictions, and distribution and delivery. This is a big job, and not just about how various agents and businesses in the industry interact with each other. It’s also about how businesses operate internally, where there is also a lot of waste.

SPRK software can also help them, said Piutti.

“If you are … a large food retailer and you have hundreds of supermarkets that you send, they even have the means to understand who is oversupplied, who is lacking.

This is not the first startup that focuses on global food. The previous startup, named Sharecy, still looks active Join LinkedIn and access Piutti’s full profile, have the same big goal:

“Our vision is a world without food waste where everyone – including future generations – has enough food and thrives.”

SPRK is now the second kick on the tin. Putting aside business results, if no one is starving, and we stop letting half of global food production rot, Piutti will know he has achieved his goal.

Get the full transcript of our conversation here.