How will AI impact spirits?

A recent conference by the drinks business looked at how AI is being used in the wine and spirits industry, and how the technology might develop.

While it may not be widespread, artificial intelligence (AI) is already being used in various industries to help with new product development, marketing and in the improvement of business operations. But how can it bring benefits to companies in the alcohol sector and does it a pose a risk to creativity and jobs?

The topic was discussed in the drinks business’ conference on AI on 17 June at London’s Science Gallery, exploring how AI is transforming the wine and spirits industry. The event was held in partnership with Preferabli, product discovery and recommendation software for wine, spirits and food.

Among the panellists for the first seminar, The Nature of Sensorial AI, were: Patrick Schmitt MW, the drinks business’ editor-in-chief; Pam Dillon, co-founder and CEO of Preferabli, New York; professor Charles Spence MA PhD, head of the Crossmodal Research Laboratory at the University of Oxford; and Chris Hoel, founder and CEO of Harper’s Club.

Spence highlighted how food and drink have a point of differentiation that extends beyond the product to the experience. Citing the example of vodka or water that are “defined by being tasteless”, he said: “How would you deliver a premium experience there? Well, if you can connect it to a sort of sonic seasoning or some other kind of multi-sensory experience that goes beyond the drink, then you’re delivering more, and you’re building a differentiated product, and more tastes converge.”

Sommelier Hoel believes music is another important element in a service setting. “It heavily influences the mood that you’re in,” he added. Spence also cited one example during a Campo Viejo wine tasting over four days where “simply changing the lighting or changing the music had a profound impact on taste”. He added: “How can we go beyond the normal range of experience to deliver extraordinary tasting experiences, something that you’ve never had before? Could you have a magical serve?” Spence said Glenmorangie has looked at extraordinary tasting experiences with whisky using ASMR (Autonomous Sensory Meridian Response) triggers, so consumers could ‘feel’ the taste of Glenmorangie.

Dillon said the world was in a “massive redefinition of what constitutes a luxurious experience”, with consumers becoming more conscious of their planet.

Another panel discussed how generative AI, which can produce text, images, or video, can be applied in the industry. Schmitt was joined by Andrew Sussman, co-founder and CTO, Preferabli; professor Oguz A Acar, professor of marketing and innovation at King’s Business School, King’s College College; and Nick Martin, CEO of Wine Owners.

On how brand owners can use generative AI to market their products, Acar noted several AI opportunities in the market. The first one is customisation and how we can “take personalisation to the next level to an unprecedented scale”. The second is creativity: “What’s interesting is when humans work with AI, the benefits can increase and make you more creative.”

Speaking about the future of AI, Acar speculated that there could be more “agentic AI – AI tools that can do a task themselves” as currently they need a lot of human input. “The second one I think we’re going to see is increased context windows… tools that can actually keep much more in their memory.”

The role that AI plays in cocktails also came under discussion, with Will Meredith, a cocktail bar veteran and director of Daisy Age Drinks, and Lucy Thomas, head waiter at London bar Lyaness. “We are always looking for ways that we can have the widest reach and be as diverse as possible,” said Thomas, when it comes to creating new cocktail menus with the whole team’s input. “We were trying to think about how we can employ AI in a way that isn’t AI writing our menu, because that will kind of take our jobs – but implementing it in a way that was interesting. And so we had our menus based on core ingredients. We had five core ingredients at the time. And basically we posted the names of these ingredients into an AI generating software that then generated the imagery for our menus and coasters.”

One example was a drink based on a mushroom, which saw AI create imagery that replicated the cloud created by an exploded atom bomb. “Every menu, we just punch in a few more ingredients and just see what photos we can create for them. It’s just a nice playful way of integrating [AI] without it taking away our [design] agency or creativity ourselves.”

Schmitt also highlighted Diageo’s recent collaboration with AI Palette, which revealed key flavour trends for 2024 including umami, spicy, tropical and ‘bloom harvest’.

Meredith noted the importance of “pushing ourselves” when it comes to looking at ingredients, flavours and drinks, at a time where “everyone has access to just about every ingredient on the planet”.

He added: “Trends are generally dictated by macrocultural events, whether it be cinema, art or fashion. Last year was Barbiemania, and everyone was doing pink drinks in the latter half of the year. Then there was the Martini, after Oppenheimer. You’ve got the Pantone colour scheme, which dictates the Starbucks slushies. Drinks trends definitely follow this kind of macrocultural movement.”

Software development

The final seminar of the day explored AI software development in the wine and spirits industry. Joining Schmitt and Dillon on the panel was Barry C Smith, professor of philosophy, and director of the Institute of Philosophy at the University of London’s School of Advanced Study, and Rob Hollands, CEO of Proofworks, which aims to help the spirits sector digitalise the supply chain and combat counterfeiting.

On whether AI could replace roles in the industry, Dillon responded by saying that the Preferabli software “is not replacing anyone. “What I see broadly in the wine and spirits industry is optimisation of production and supply chain. There are elements of software that are replacing some jobs but at the same time creating others, but in the narrow field of wine and spirits software machine learning, there is no eradication of jobs. This is all amplification and expansion.”

Hollands added: “From a consumer perspective, not everyone has access to go to a whisky tasting or visit a distillery or be guided on their wine journey. So digital tools can be used to do that sort of replacing and scaling out, and making them more accessible to bigger audiences. In terms of actual distilleries themselves, we are optimising the process there. But it’s to free up distillers to do things that they’re great at: creating new recipes, innovating
in other ways.”

Proponents of AI maintain the spirits industry can enhance operations efficiency, innovate with new products and recipes, improve customer experiences, and stay competitive in a dynamic market.