The latest news, trends, analysis, interviews and podcasts from the global food and beverage industry

With AI becoming increasingly widespread across all industries, Dr Stephanie Brooks, head of research and innovation at Foods Connected, examines how food and beverage businesses can leverage this technology to achieve their sustainability goals – and why, despite its potential, many companies are failing to harness AI effectively.

It is so exciting to see that food and drink businesses have, by and large, embraced the potential of artificial intelligence (AI). They are deploying AI on production lines to improve product quality, investing in customer-facing chatbots to relieve pressure on CX teams and installing in-store cameras that automate availability.

In fact, nearly 70% of agri-food businesses have implemented AI at some stage of their supply chain (or are planning to do so) according to Foods Connected’s exclusive survey of 500+ senior industry professionals. Food manufacturers lead the way here too – 49% are making use of AI and machine learning technologies vs 36% in food retail. But unfortunately, it isn’t all good news. Worryingly, the research uncovered that food firms are failing to utilise AI when it comes to one of the sector’s biggest priorities: sustainability.

That is a huge missed opportunity. It bears repeating that sustainability is no longer a nice-to-have; it is a commercial imperative in agri-food. For one, we know how much consumers care about the environmental impact of the food and drink they consume and the impact this is having on how, where and what they shop.

A recent consumer poll by data firm Savanta found that 72% of UK customers say their purchases are now influenced by a desire to shop and eat more sustainably. They are willing to pay a premium for products that align with these values, too. According to McKinsey, 60% of global shoppers would dig a little deeper in their pockets for eco-friendly packaging, for example, with products making ESG-related claims averaging 28% cumulative growth from 2018 to 2023, versus 20% for products that made no such claims.

Second, though are the regulatory pressures that food firms are facing. From the EU’s Regulation on Deforestation-free Products (EUDR) to Extended Producer Responsibility in the UK, the need for accurate, granular oversight of your entire supply chain in order to meet these emerging legal requirements on sustainability has never been more pertinent.

How to use AI to back up sustainability claims

Now, AI isn’t some magic bullet, but it does offer exciting new ways to both accelerate progress toward ESG ambitions and communicate that progress in a more efficient, standardised way, alleviating what could be a huge headache for the industry.

To begin with, AI offers the opportunity to credibly cement your product as a sustainable option, validating on-pack claims with irrefutable data evidence – and build trust with a growing group of eco-conscious consumers. For the 30% of consumers who told Savanta they currently struggle to identify if a product is sustainable or not, AI and the data behind it allow manufacturers to swap vague language for specific substantiated claims that consumers can verify.

This closes the all-important credibility gap at a time when consumers are crying out for products that can back up their claims. AI can also be used for smart utilities management, be it helping to capture energy from renewable sources more effectively, or optimisation of energy and water use to reduce environmental impact.

How AI can help reduce food waste

Then there are AI models that use predictive analytics models to pinpoint spoilage. These tools can harness data on temperature, humidity and chemical composition, among other metrics, to identify the moment when food is likely to spoil, equipping manufacturers with the ability to reduce waste (and cost) and even provide retail customers with dynamic expiration dates that can slash household-level waste, though to be behind some 8-10% of global emissions alone.

It can be used in agriculture as a tool with which to improve animal welfare too – another key ESG focus for UK consumers. 68% say animal wellness is either very or extremely important to their purchasing decisions, according to 2024 research by public health body NSF and AI, once again, allows manufacturers to demonstrate the veracity of their claims. Agri-food businesses can use AI surveillance tools, for example, to closely track livestock behaviour and quickly spot any signs of disease, stress or malnourishment.

Camera Vision systems can even analyse hoof health in cattle using water baths to spot lameness – a major welfare and economic concern for farmers. And then there’s the macro-level impact. With large food firms facing a growing raft of ESG reporting obligations, AI can provide accurate, granular and shareable data at each stage of their supply chain, without having to grapple with reams of often varying and patchy datasets provided by their vast network of suppliers. Suffice to say, AI isn’t a panacea, but it does have enormous potential when it comes to sustainability. So why aren’t food and drink businesses taking advantage?

Well, the research uncovered a few big hurdles that could be responsible. First up, we lack some of the key skills we need in industry to really identify and unlock the value that AI could bring to sustainability efforts.

In particular, we’re grappling with a shortage of data scientists, a new and emerging job market of experts who are equipped to work with and within the industry. Though there are data scientists, typically from an academic background, out there, the general lack of application of these skills together with experience and technical knowledge of agri-food experts to real-world settings often means solutions sometimes miss the mark and fail to gain traction. Instead, we need access to data scientists who can better bridge the gap between academic research and industry application to make AI solutions that are effective and applicable in practical settings – and therefore more appealing. Thankfully, we’re already seeing some fantastic initiatives emerge that could help plug this gap.

What needs to be done

Momentum One Zero, spearheaded by Queen’s University Belfast, for example, is a global innovation centre that brings together collaborative experts and delivers training for many more, with the aim of applying their skills and knowledge within agri-food organisations.

The effects of this won’t be felt overnight, but is a step in the right direction, one that will hopefully trickle down and start changing mindsets at big organisations. Even with the right expertise in place, of course, the barriers don’t stop. The food and drink sector – known for its small margins – is also notoriously cautious in adopting new technologies.

Each investment faces intense scrutiny, and there’s an understandable reluctance, particularly in today’s tough economic climate, to take a risk without proven efficacy. This means that while AI has been readily embraced in areas where it is well established, Food Connected’s research found it was primarily being used to enhance quality control, quality assurance and inventory control. For example, emerging areas of application, such as sustainability, are being met with greater hesitation.

The reality, though, is that the first movers in this space will reap the biggest rewards. The first suppliers to verify on-pack claims using AI will see the biggest uptick in engagement from both retail customers and consumers, while those following months, even years later, will be seen as firmly behind the curve.

The same goes for manufacturers that invest in predictive analytics tools to reduce waste or agricultural businesses that can credibly promise higher welfare livestock. Those that spearhead the use of technology in this space may be undertaking a bigger risk, but they’ll also differentiate themselves from the competition in a far more meaningful way.

In short, when it comes to both AI and sustainability, the food and drink sector is at a crossroads.