Written by Aidan J. Connolly (Cainthus and AgriTech Capital) and Saoirse Boret (University of Amsterdam).
Artificial Intelligence
Artificial intelligence (AI) refers to the collection of data from sensors and its conversion to comprehensible information. AI machines can mimic human cognitive functions such as learning and problem solving, and interpret information more efficiently than humans, reducing their need to be involved. As it is being developed, it is clear that AI can also be self-learning, and progress beyond human abilities. The use of AI to advance food production is being widely considered as the world progresses towards a more modern way of living, and expectations of speed, efficiency as well as sustainable advancements are ever-increasing alongside the rapidly growing population.



Controversy
With all positive aspects of AI some see it as a technology with the goal of taking over human jobs, and that creates controversy. The fear of the unknown is creating a pushback against the use of AI in many businesses. Additionally, AI requires skilled IT professionals, which are high in demand and difficult to recruit. Clearly there are costs to retraining programs to adapt to the change in skills required. Finally, the cost of implementing and maintaining AI is very high, which may limit the opportunities for smaller or start-up business to compete with already established larger ones. This may lead to a reduction the numbers of smaller businesses, who would likely not be able to compete with the investments required and would be forced out of the market.

Downsides such as these could possibly slow down the speed with which AI transforms food production, but given the absolute power unleashed by AI in a post pandemic food world, it is unlikely to be more than a speed bump on its eventual universal acceptance. The partnership of EIT Food with Microsoft and Danone to “select and accelerate 6 Agri-food startups specialising in [AI]” is an example of an efficient way for larger companies to boost the awareness and understanding of AI’s potential benefits, and increase a drive for research and development in the area.

Artificial intelligence is reimaging the future of food, agribusiness and the connection with a changing consumer and the current pandemic has accelerated the supply chains willingness to embrace its full potential.

Following are six key areas in the food industry where AI has been introduced, has accelerated growth, or even changed the way in which companies operate.

It is clear that employing AI-based solutions such as UAVs would benefit farmers in more ways than one, as they would be saving money as well as complying with the laws put in place to protect consumer health, and ultimately lead more sustainable practices
1) Processing
As farming continues its journey of modernisation in a quest to feed the burgeoning world population, progressive farmers are including AI technology as part of their management mix. The processing of food is a labor intensive business, but one where AI can maximize output and reduce waste, by replacing people on the line whose only jobs are to identify items unsuitable for processing. Decision making of this type at speed requires the senses of sight, smell, and their adaptability to adapt to changing circumstances. AI brings even more to the table through augmented vision, analyzing data streams either unavailable through human senses, or where the quantities of data are overwhelming.

Organizations such as TOMRA have already begun to incorporate AI technology into their production processes, including innovative sensor based sorting machines, detecting and removing any types of foreign materials from their lines of produce, reacting to changes in moisture levels, colors, smells, and tastes of foods. They claim the recovery of “5-10 per cent of produce through higher yields and better utilization” equivalent to “25,000 trucks of potatoes per year”.

Electric Noses
Along the same lines is Aromyx, who use AI to quantify taste and smell to increase efficiency in production in order to create “a new quantitative and visual standard to represent... senses of taste and smell as actionable data”. Another similar technology are electric noses, which as the name suggests, can be used as replacements to human noses to distinguish various foods’ odors and aromas using AI sensors.

Detection
Another example is how Japan's Kewpie use Google's Tensorflow AI for the detection of defective ingredients during processing. It was originally used for the sorting of foods, and gradually developed into an anomaly detector, which could then be used for unsupervised learning, saving both a large amount of time and money. While focused on diced potatoes at this point in time, the company plans to “expand to eggs, grains, and so many others”.

Gamaya has targeted agrochemicals to lower environmental impacts and costs during food production, through the use of AI to create maps which analyze various crops’ phenological and physiological traits. This can then be used to “provide targeted and tailored recommendations for the optimum management and treatment of [their] land and crops”. More examples of AI based systems include Greeva and Raytec Vision for sorting fruit, Sesotec eliminating contaminants, A myriad of other agri-tech startups are focused on using AI to detect early warning signs of poor health in crops which may have otherwise been overlooked, which can further reduce the waste in food production, on top of increased transparency. For example, CAMP3, a small American company, uses AI to spot plant spots and diseases at an early stage, and reduce the chance of contamination further along the production line. Similarly, AgEYE Technologies is another US start-up which strives to detect pathogens and prevent contaminations using AI. HelioPas AI specialises in the early identification of droughts, which further helps farmers optimise their yields.

Monitorisation
AI has also been developed for the analysis and monitorisation of crop fields, through the use of unmanned aerial vehicles (UAVs), such as AgEagle aerial systems. These are used to keep farmers informed of any areas which might be facing issues and need assistance. This practice saves a lot of time for farmers, who can then focus on farming on a wider scale. Consisting of “multispectral imaging sensors, GPS map creation through onboard cameras, heavy payload transportation, and livestock monitoring with thermal-imaging camera-equipped drones.” This appeals to the business aspect of agriculture, as this would save both time and money. In addition to this, the use of UAVs would be more ecologically friendly as its’ goal is to increase efficiency, therefore reducing the need for spraying pesticides on large pieces of land, as well as reducing waste by noticing anomalies at early enough stage.

Ultimately, this can lead to better consumer health, falling in line with the increased restrictions in today’s agricultural farming for health and sustainability purposes. This intelligence within the drones also aids in reducing the emissions of fossil fuels, and it signifies less traffic pollution, as it looks to reduce farmer activity. It is clear that employing AI-based solutions such as UAVs would benefit farmers in more ways than one, as they would be saving money as well as complying with the laws put in place to protect consumer health, and ultimately lead more sustainable practices. Increased monitorisation of crops would likely increase consumer confidence in the food products. Demand for sustainably produced food is moving from niche to mainstream, with consumers increasingly conscious of the provenance of their food and ingredients, which can be achieved through this increased traceability.

Grow Crops
Many of these field applications have been covered in the Crop Farmers digital dilemma. One Soil is a satellite-centered AI solution which provides detailed analysis of weather patterns using data collected from the EU-based Copernicus program. Seedo, an Israeli start-up, is a high-tech company which uses AI to provide “fully automated and controlled indoor growing machines”, to allow for the optimisation of land use, so people can grow crops in their very homes. They have begun with medicinal cannabis, providing optimal light and general conditions for its growth. Although many uses of AI relate to crop farming, but it has also proved to be effective in cattle farming. Companies such as Irish-based Cainthus makes use of computer vision and intelligent analysis, to shows farmers a clear view of their Dairy cows and general barn environment and predict what is likely to happen next, based on the scrutiny and interpretation of cattle behavioural patterns.

So, for example, it can detect early warning signs of poor health which may have otherwise been overlooked, likely leading to a decrease in waste during the production, as well as reducing the need for human labour, subsequently raising the efficiency levels in the farm. This ultimately strives to increase sustainability within the practice, as it focuses on increasing food production from pre-existing animals, which may lessen the need for numbers of cows or increase the productivity, reducing levels of methane gases in the environment. This technology also ensures better animal welfare, as cow health and comfort are being constantly and closely monitored. Alongside animal welfare, traceability is also improved through the use of AI, and consumers are better informed as to where exactly the food they are eating comes from.

This is especially important now during times of Covid-19, with an increase in expectations for sanitary production, as people have more time to cook, as well as research where their food is coming from and the impacts on their surroundings its production may cause. The removal of the chance of human error is also likely to increase consumer confidence in the company and encourage higher beef consumption, which is the ultimate goal for the business.

More recently the company added improved facial recognition abilities to account for the mandatory use of a mask and new body temperature detection
2) Food Safety
Reducing the presence of pathogens and detection of toxins in food production is a key avenue for AI. The Luminous Group, a Newcastle-based software firm, is developing AI to help prevent outbreaks of pathogens in food manufacturing plants, limiting consumer illness or recalls. Additionally, AI offers the opportunity to increase traceability and consequently, consumer confidence. For example, a KanKan subsidiary consisting of AI-enabled cameras in Shanghai’s municipal health agency checks that workers are complying with the safety regulations. This algorithm-based machine learning technology includes facial and object recognition, and “sets the foundation (...) to potentially triple [their] business with the city of Shanghai”. More recently the company added improved facial recognition abilities to account for the mandatory use of a mask, and new body temperature detection, in line with effects of Covid-19, as detecting increased body temperatures could help in the early detection of a Covid case. This everchanging project shows an ability to constantly grow and develop, a flexibility required today in the world of technology.

Clean Food Prepping
Dragontail Systems Limited is another software company which has taken up many of the above-mentioned new measures during the fight with Covid. Additionally, Japanese company Fujitsu has developed an AI-based model which is used to monitor hand washing in food kitchens following strict regulations set by the Japanese health ministry. This technology will reduce the need for visual checks (critical during Covid) where food safety must be increased. iCertainty have been working for a decade or more with Disney on clean food prepping protocols which Myra Analytics strives to make compliance in food chains easier to track. Next generation sequencing (NGS) is also another attempt made by food safety associations to increase the accuracy and speed in which any threats to food safety are identified and dealt with in a production chain.

Cleaning In Place
Thirdly, AI can be used for Cleaning In Place (CIP) projects, which aim to use AI to clean production systems for cheaper, and using more environmentally friendly methods. In Germany, a project by Industrial Community Research strives to "develop a self-learning automation system for resource-efficient cleaning processes”. This makes up a cleaning process without a need for the disassembly of equipment. This could cut labor costs and time spent on it, as well as increasing the safety of food production in the plant in question by removing the opportunity for human errors. The University of Nottingham has also been working to construct a Self-Optimising Clean-in Place system, which uses AI “to monitor the amount of food and microbial debris in the equipment”. Centaur takes a different approach to focus on safe grain storage, using AI powered sensors and management systems.

Covid-19 has accelerated the applications of technology to replace human labor and while smart device Food apps, drone and robot delivery, and driverless vehicles all provide new ways to get information and food to the consumer, all of them depend upon AI
3) Supply Chain Efficiencies
UberEATS, which now sits at “a $6 billion bookings run rate, growing over 200 percent” riding the wave of popularity of food delivery, is now incorporating AI to make “recommendations for restaurants and menu items, optimise deliveries”, as well as looking into the use of drones. They use Michelangelo, a machine learning platform, for various different tasks. For example, this can predict meal estimated time of delivery (ETD) to reduce waste and improve efficiency throughout its delivery process. While this application is post-food production once food has been produced there are many more ways to implement this up and down the chain. Covid-19 has accelerated the applications of technology to replace human labor and while smart device Food apps, drone and robot delivery, and driverless vehicles all provide new ways to get information and food to the consumer, all of them depend upon AI.

Dallas-based technology company Symphony RetailAI uses AI in the food supply chain to “boost productivity, and greatly improve the accuracy of information for better decisions”. Innovative uses of AI are crucial in moving towards reducing the quantity of food wasted in order to feed the growing world population as efficiently as possible, as well as falling in line with increasingly specific consumer demands and expectations. Shelf Engine offers AI to remove human error from the purchasing function and make more informed decisions about order sizes and types, in hundreds of US stores has saved thousands of dollars in food waste. Wasteless have been even more aggressive, allowing retailers to use dynamic pricing to discount produce before it goes past its sell by date. Ocado, a British online supermarket, uses AI to predict the quantities of food that their consumers realistically want, as to not produce excess foods. They also use this technology in their warehouses, to guarantee that the food they deliver arrives on time and is fresh, in an attempt to battle the prominent problem that is food wastage.

4) Predicting Consumer Trends and Patterns
AI allows companies to stay competitive within the market, by adapting based on different popular waves of various trends, making predictions about the market. Michelangelo is again one of the most innovative examples of predicting trends, it’s detailed method being explained in greater detail on the hotlink. Tastewise is US startup using AI-based intelligence for food retail. Their data collected includes “up-to-the-minute industry insights, predictions, and emerging food trends based on analysis of billions of social media posts and photos, US restaurant menus, reviews, and recipes”. McCormick & Company and IBM is funding research and development of “flavor and food product development” through the use of AI.

As previously mentioned, Aromyx enables “clear, reproducible digital representations of smell and taste”. Their website claims to “accelerate new product development”, speeding the process of adapting to new trends or predicting future ones. Bite.AI is another player in the space. Carlsberg Research Laboratory has also jumped on the bandwagon, announcing its participation in ‘The Beer Fingerprinting Project’, including a large investment in a “research study with the purpose of measuring and sensing flavours and aromas in beer”, to further cater to their consumers’ demands, satisfy them and increase sales.

5) Restaurants
The future of restaurants is in peril following this year’s Covid-19 outbreak. The explosion of online-based food delivery systems has decreased the focus on the physical experience of restaurants. For example, chat boxes can allow communication with your favourite restaurant without leaving the comfort of your home, all powered by AI. Voice search is another tool useful to allow people to place restaurant orders simply by talking to their screen.

AI analytical solutions such as these lead to better consumer experiences and likely to increase sales for restaurants due to the ease with which food orders can be placed. With the previously mentioned example of UberEats, AI is used to increase efficiency and lower costs during the process of food delivery, encouraging restaurants to partner with these companies to ensure the delivery of their food.

Automated customer service and segmentation will likely lead to increased accuracy in “creating reports, placing orders, dispatching crews, and formulating new tasks” in a restaurant. For reasons of added safety and precaution, restaurants are also likely to have to adapt new health-cautious AI technology systems in their kitchens, as mentioned in paragraph 3, including Dragontail and Fujitsu.

Artificial intelligence is reimaging the future of food, agribusiness and the connection with a changing consumer and the current pandemic has accelerated the supply chains willingness to embrace its full potential
6) Designing better foods
Food is health has been a mantra for many for a long time, but now with a greater understanding of both human, plant and animal genomes it is becoming a reality. Changes in consumer preferences are creating opportunities for AI in food; an example is the growing demand for plant-based alternatives to meat protein, as the world moves towards precision nutrition. Challenges such as achieving consumer acceptable taste and texture qualities has led to creative AI applications.

NotCo is a plant-based start-up company located in Chile which has been developing its own software company ‘Giuseppe’ a tool used to “predict how to make plant-based materials taste like animal-based products”. This AI system uses machine learning and genomics to find similarities in plant and animal protein products. The goal of the company is to move towards healthier and more sustainable food sources at lower costs.

Health Benefits
Brightseed, a bioscience tech company, makes use of AI to further understand the health benefits of plant crops, in order to better human health with their food produce. This includes the acknowledgement of any helpful antioxidants which may be present in their foods, ensuring that these are carried throughout all the stages of production and delivery. For milk, meat and eggs remote observation through wearable and environmental sensors but cameras including those from Cainthus are scalable and address productivity, animal welfare and concerns about the environment also.

Food Pairing
Gastrograph is a food technology app which collects data from its consumers and uses AI to convert it into new, comprehensible, and helpful data. It helps people figure out what they are tasting and helps provide tailored recommendations for what they should try next. This type of application is very appealing to consumers as it is easy to use and is very personalized. Foodpairing is another platform which uses AI to create various combinations of foods which can potentially be used together to create meals, an app aimed to target chefs and bartenders. The application of artificial intelligence can be witnessed right across the food production spectrum. FlavorWiki uses machine learning from market trends to design better products.

Leading taste and nutrition company Kerry have developed Trendspotter – a tool that automatically reads and processes millions of consumer-generated social media posts, extracting food items and cataloguing food-related combinations. Using an algorithm Trendspotter calculates and predicts the ingredients, flavors, foods and products most likely to match consumer’s evolving tastes. These insights are used in tandem with other industry-leading primary research capabilities to make targeted and on-trend recommendations.

What do you think? Can Artificial intelligence be the food system’s saviour? Please share your thoughts and ideas in the comments below.

Thanks to Saoirse Boret of the University of Amsterdam, for her research and writing for this article.

Aidan Connolly
Aidan Connolly is CEO at Cainthus and President of AgriTech Capital. Connolly is a senior Executive with 28 years of successful global business leadership experience in the food & agribusiness sector. He has expertise in strategic management, sales, financial management, innovation, talent development and management, nutrition, branding, and international marketing. He has direct experience in greenfield start-up operations, high growth, turnaround, challenging economic environments, and a wide range of political and economic systems. His substantial leadership experience is in all areas of organizations, from strategy to operations to production, developing sales program, and building and developing cohesive teams.