Microsoft Has Built AI That Can Make Perfect Whiskey For A Region's Taste Palette
Whiskey has a variety of tastes, depending on how and what you brew it from, what you store it in, and for how long. But now, Microsoft of all things have helped produce a new brand of whiskey. And they didn¡¯t even have a master distiller.
Whiskey has a variety of tastes, depending on how and what you brew it from, what you store it in, and for how long. But now, Microsoft of all things have helped produce a new brand of whiskey. They didn't even have a master distiller, so they created their own.
Microsoft recently offered the services its Azure cloud platform to Swedish distillery company Mackmyra Whisky, and technology brand Fourkind. What do those three companies have in common? Apparently a desire to create an AI that can brew the fine alcohol.
The team used a combination of old-fashioned techniques like cask-aging, with a high-tech AI to design new flavours of whiskey. The distillery used Microsoft's Azure cloud platform to power its machine learning model. They trained it on data regarding the distillery's existing recipes, sales data, and customer preferences. The idea was to have the idea cross-reference all of these and come up with a recipe that would be a hit with consumers.
As Microsoft said in a statement, "With this dataset the AI can generate more than 70 million recipes that it predicts will be popular, and of the highest quality based on what kind of cask types there are in the warehouse."
Good whiskey normally spends a minimum of three years distilling in a cask, with a 12-year minimum for scotch. The type of wood used can change the flavour greatly, but there are plenty of other subtle details that can influence it too. That means there are potentially infinite ways to brew it.
The only way to test these out however is to make your recipe, allow it at least three years to age, and then try it out. And if the recipe is bad, you've just wasted all that time waiting to find out.
What this AI does then is reduce the risk of the waiting period. By comparing successful recipes with sales data and what customers say they like, the algorithm can come up with a new recipe that has a higher likelihood of turning out well. And since this is all technically being simulated within the AI, it's essentially hallucinating these recipes from the number chaos.
"We always strive to challenge the traditions in the very traditional whisky trade, and that's something we can really do now with the help of AI,"says Mackmyra Whisky's Master Distiller Angela D'Orazio. "For me as a Master Blender, it is a great achievement to be able to say that I'm now also a mentor for the first ever created AI whisky in the world."
It was only a matter of time before this happened after all. Give humans any technology and they're going to try and find a way to get high off it.
