View all on-demand sessions from the Intelligent Security Summit here.
Computers have a bad reputation when it comes to saving the planet. Cryptocurrencies, due to the highly inefficient technology involved, have consumed as much electricity as the entire country Sweden. Elon Musk has repeatedly warned about one terminator-like apocalypse likely caused by artificial intelligence (AI). And yet, like any other tool, AI has enormous potential to be good for the planet – and this future is not as far off as it seems.
Today, let’s explore one aspect of this potential: reducing carbon emissions from food-related systems. According to Nature, these account for a third of total emissions; the growing world population points to the increasing significance of this factor with time.
Computers are great at keeping track of countless factors and adjusting output without human intervention. There are at least two food system-related tasks for which this is well suited: reducing food waste and promoting the consumption of food that is better for the environment. Let’s look at each of them in detail.
Contents
Preventing food waste at home
According to the USDA21% of the food consumers bring home is wasted and another 10% is thrown away in the supermarket/warehouse. Let’s look at the root causes of this waste.
An important factor is that consumers do not know what to do with the food that caught their eye in the store. Maybe it was for sale; perhaps part of the item was used for a recipe and the remnants don’t offer a good way forward. When there’s no plan — no recipe to make with a particular grocery item — the chances of it going to waste increase. This is especially true for items with a short shelf life, such as vegetables and proteins.
But what if the grocery shopping paradigm shifts from focusing on individual grocery items to focusing on recipes? Each item in the fridge would then have a “plan” around it; as long as the recipes are what the family wants, the items will all be eaten.
This paradigm, combined with AI that zooms in on family food preferences and recommends recipes that any family would enjoy, has been quite powerful. Recipe-based shopping is a thing Instacart and Amazon embrace too; there’s no reason brick-and-mortar supermarkets can’t do the same.
In addition, rather than treating recipes as standalone recipes, supermarkets should think about how consumers can reuse ingredients in different recipes for the week. For example, if a recipe in the customer’s cart calls for parsley as a garnish, an additional salad recipe might use the rest of the parsley bunch. This saves customers money and reduces the chance of unused parsley going to waste.
This task – combining complementary recipes to make the best use of leftovers – is perfect for AI.
Less food waste in the supermarket
A lot of waste in the supermarket and the warehouse is the result of overstocking. Despite supply chain systems being fully automated and market incentives to improve, the USDA still estimates retail losses at 10%. Consumer behavior is pretty hard to predict as long as the business is based on consumers browsing virtual or physical aisles and choosing the grocery items they want.
What if this model was reversed? What if consumers don’t directly choose the grocery items or even recipes they want; what if they declare their broad food preferences and an agent acting on their behalf (a human or an AI) does the shopping for them? Provided that this agent represents consumer needs well, the agent can also be notified of stock levels at the retailer; they could then make substitutions that would not affect consumer satisfaction but prevent spoilage.
Besides the obvious benefit to the planet, reducing waste makes for a more profitable business and allows certain savings to be passed on to consumers. When the margins of typical supermarkets are in some numbersthese savings add up – especially in an inflationary environment.
Food that is better for the environment
Following a low carbon footprint diet is a surprisingly counterintuitive task for humans. According to Our World in Data, local food is often not better than food shipped from distant continents. Biological food often has a higher greenhouse gas footprint. Even reducing packaging isn’t the right factor to pay attention to: it’s a Small piece of a food’s environmental impact and often extends shelf life, reducing waste.
It is too difficult to keep up with the latest insights into what is actually good and what is bad, and the research is evolving rapidly. So the cognitive load persistence is too high, even for those consumers who care about sustainable eating.
Wouldn’t it be great if there was an autopilot? Something like the ESG investment funds who do the work for you, but in the food world? Something that would help you do the right thing and send you a quarterly report on how much better you did than the average Joe?
Unlike investing, where you can be as hands-off as you like, this doesn’t work so easily with food. In addition to caring about the sustainability properties of your food, you also care deeply about its taste, allergens, macronutrient content, portion size, and a host of other factors. Unless you’re vegan, there are plenty of vegan meal options you wouldn’t like, and plenty of vegan, vegetarian, and low-carbon omnivore options you’d like.
Understanding all customer needs and adapting recommendations based on a feedback loop (using structured, explicit feedback) is an important factor here.
Imagine a world where an autopilot for healthy and sustainable eating exists. If this autopilot knows each consumer well, it can confidently guide some of them towards more sustainable food – by swapping a beef-based recipe for a chicken recipe or introducing a vegetable meal to someone who normally has a strong appetite for meat. tends. AI plays a central role in these nudges because each customer’s preference is unique; and because collecting feedback at scale and adjusting recommendations based on it is key to achieving all objectives.
This concept of micro nudges is very relevant. With sustainable options in the shopping experience along with social proof, can help traditional “browse-the-aisles” retailers help consumers make the right choices. For digital retailers, more knowledge about each customer can help optimize relevance to sustainability. In the optimal case, these two variables need not compete.
AI as a force for good
As we’ve seen here, AI-based systems can help reduce greenhouse gas emissions in two ways: by reducing food waste and by encouraging consumers to eat more sustainable foods. Each of these factors could have a major impact on the planet over the next ten years.
Alex Weinstein is CDO at Hungryroot.
Data decision makers
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people who do data work, can share data-related insights and innovation.
To read about advanced ideas and up-to-date information, best practices and the future of data and data technology, join DataDecisionMakers.
You might even consider contributing an article yourself!
Read more from DataDecisionMakers
Janice has been with businesskinda for 5 years, writing copy for client websites, blog posts, EDMs and other mediums to engage readers and encourage action. By collaborating with clients, our SEO manager and the wider businesskinda team, Janice seeks to understand an audience before creating memorable, persuasive copy.