The wine industry favours tradition. The first traces of wine production were near Tbilisi in Georgia, around 6,000 BCE, but over the centuries, the industry has adapted to technological advances. The glass bottle was uncommon until the 17th century, though the first ever glass bottle dates back to Roman times. The wine industry is worth around $340 billion globally, employing over a million people directly. The EU wine sector alone supports around 3 million jobs, directly and indirectly, and there are over 100,000 different wineries around the world. Modern wineries use plenty of technology, from drones with infrared cameras to monitor vineyards, to grape sorting machines and smart labels to combat counterfeit wine labels.
The industry is starting to embrace artificial intelligence (AI). An experiment at the University of Geneva in 2023 showed that AI could be used to categorise the chemical structure of specific bottles of wine, being able to identify 80 red wines from different Bordeaux estates and vintages with perfect accuracy. This technique has potential in combating counterfeiting of expensive wine, an all too common problem in the industry. It has been estimated that as much as 20% of all wine sold globally may be deliberately mislabelled, so anything that can help in that particular battle will be welcomed by wine producers.
Another issue with wine is when the bottle is “corked” i.e. is faulty and tastes bad. This problem has many causes, but issues with the cork allowing oxygen into the wine bottle are common, and happen in as much as 3% of all wines. Technology from a company called Cork Supply maps the internal structure of a natural cork with a scanner, and then uses machine learning to identify anomalies in the structure compared to a database of regular cork structures.
Another area where AI is being used is in personal recommendations for wine buyers. A company called Preferabli uses AI to match the broad wine preferences of a shopper to suitable wines. The system prompts shoppers to express preferences such as a white wine from a certain region that is crisp and light. It then accesses a large database of wines where many characteristics have been noted by wine experts, and recommends particular wines to the shopper. It can also suggest food pairings for the wine. The machine learning analyses the taste characteristics of millions of products to drive its recommendations. Rival company Tastry uses a similar machine learning approach to deliver customised preferences. The largest wine marketplace, Vivino, has a database of over a million wines and has launched a wine recommendation service based on customers’ previous preferences and searches, utilising AI models.
Vineyard health can be monitored by aerial sensors from companies like the Canadian firm VineView. These measure water content in leaves, using AI to stitch together images from drones and to analyse those composite images. One wine maker even used ChatGPT to help design a new wine, from suggesting the blend of grapes to be used to designing the label, its name and the choice of bottle. It can be bought from their website. The process of fermentation is also seeing internet-of-things technology being used, with sensors embedded in fermentation tanks providing real-time insights and alerts, for example, if a vat is overheating. Machine learning models can be used to predict potential quality issues before they occur.
What about assessing the quality of wine? This has long been the preserve of wine experts, an area shrouded in mystique. The Wine and Spirit Education Trust has a series of exams that aspiring sommeliers can take, resulting in a range of qualifications, the most prestigious of which is “Master of Wine”. Just 425 people in the world have so far achieved this level, as of early 2025. The regular ChatGPT chatbot has passed the theory exam for the WSET exams at a high level, and also the related Advanced Sommelier exam (scoring 77% in 2023), and so it can also make good general recommendations on wine.
We tend to associate leading-edge technologies with early adopter industries like investment banking and life sciences, but the example of the wine industry embracing AI in several ways shows that it can have an impact even in quite traditional industries. In the future, we can expect to see more technology, some of it AI-based, employed in vineyard management, in wine making such as the fermentation process, in addition to more obvious areas such as marketing and sales. The use of AI to help improve cork reliability, to potentially trace the lineage of individual bottles of wine to combat fraud, as well as provide tailored wine recommendations to individuals, are examples of an industry that can adapt. We can expect to see wine to be a blend of tradition and modernity in the years to come.
