Tag Archives: machine learning

Are machine-written wine reviews closer than ever?

AI wine reviews
“Come on, already. What’s taking so long? It’s only a bunch of crummy chardonnay reviews.”

Dartmouth researcher: “We couldn’t tell them apart from human reviews”

A group of Dartmouth researchers have apparently taken the next step in machine-written wine reviews. Their study, published earlier this year in the International Journal of Research in Marketing, takes the work we’ve seen with artificial intelligence and brings it to a terrifying new place: The results “show strong support for the assertion that machines can write reviews that are indistinguishable from those written by experts.”

Keith Carlson, a doctoral research fellow at Dartmouth’s Tuck School of Business, didn’t say that the algorithm he and four others developed would show up as part of Wine Spectator tasting notes anytime soon. Some work still needs to be done, including further research into how similar the machine reviews are to human reviews.

Which is what I noticed: The examples were close, but not quite there – one machine rose review mentioned vanilla, which isn’t all that common in rose.

Still, Carlson told me, advancements in machine neural networks allowed the group to apparently come closer than anyone has before in making AI-generated wine reviews practical. Their algorithm used thousands of tasting notes from wine magazines to come up with a way that a machine could “teach” itself to write reviews.

Take that as you will. Regular readers will know how I feel. Wine has enough problems without some marketing guru — or, even worse, some media type – using AI reviews to flog product on the cheap. If this is practical, I can see AI reviews trolling sites like Vivino and Cellar Tracker, making consumers think the wines are more popular than they are – or taste a certain way when they may not.

That the study was published in a marketing journal is especially bothersome, as if the only use for technology is to sell stuff. And the Dartmouth news release even used the word “disrupt,” which is tech speak for “this will make us a gazillion dollars and aren’t we smarter than everyone else.”

Carlson emphasized that the last thing the group wants is for AI reviews to replace people. Rather, he says, they can be used to make wine reviewing easier, as a starting point for human reviewers. Perhaps. But I don’t know that the machine’s cabernet sauvignon review – dry, with blackberry fruit – is going to tell me anything new about cabernet.

But what do I know? I’m not out to disrupt anything.

More about AI wine reviews:
• “Crisp and fresh:” AI wine writing strikes again
• Artificial intelligence and wine recommendations
• AI wine writing: Maybe it’s not around the corner after all

“Crisp and fresh:” AI wine writing strikes again

ai wine writingWill AI wine writing eventually make wine tasting irrelevant?

This is a mineral-driven wine that’s crisp and fresh, with a flinty edge. It is very tangy, with zesty citrus, giving a bright character. It needs time to mature, so wait until late vintage.

That’s a review of one of my favorite $10 wines – the Chateau Bonnet Blanc, a white French Bordeaux. But I didn’t write it, and neither, technically, did any other human wine critic.

Instead, it was written by an artificial intelligence – the Wine Review Generator created by long-time wine industry executive Michael Brill. Brill, who also does tech, software, and AI consulting, wanted to find out if he could “teach” a machine to write tasting notes.

And, for the most part, that’s what he did.

Brill left a comment about last week’s blog post about the future of AI wine writing. That led to our phone conversation this week, where Brill said improved technology has made it possible to create the Chateau Bonnet review with a minimal amount of human programming. All you need, he said, is a database of wine terms, wine regions, grape varieties, and so forth. That information, combined with advances in neural network research that have helped scientists better understand how to program machines to “think,” led to the review software and to the Bonnet review.

In this, Brill said, a machine’s ability to “write” longer and more coherent sentences has improved tremendously. Before, he explained, an AI story might be half readable and half nonsense, and the most it could create was a 10-word sentence. Today, those numbers are 90 and 10 percent, and it can write a readable 10-sentence paragraph.

How the machine does this, needless to say, is incredibly complicated. It makes predictions about what comes next in a sentence based on the words that came before, a process that is much more like writing than previous AI efforts; those were more like filling in a template. Here, the AI has “learned” that a mineral-driven wine is crisp and fresh, and not oaky and flabby, so it picks the former phrase to follow mineral-driven instead of the latter.

Which is why the Chateau Bonnet Blanc effort is not a bad tasting note. It’s mostly accurate (save for the bit about aging) and it conforms to the rules of grammar and the sensibilities of wine. That the machine wrote the review without tasting the wine is impressive, and knowing only the cost and some characteristics, is impressive. And more than a little spooky.

And not just because an AI is cheaper to hire than I am. Brill said advances in machine writing could eventually make product reviews useless. Some of that happens today on Amazon, where it’s not uncommon to see badly written AI reviews praising a product. But the situation could get even worse as AI writing improves.

A top-notch AI could flood Amazon with machine-generated positive (or even negative) reviews, with the resulting effect on sales. Or it might be possible for one restaurant to force another out of business with an AI-written campaign on Yelp.

And who would know the difference?