A team of researchers has put together a prototype app that homogenizes many reviews of a wine to produce a new one.
The researchers — from Dartmouth University — pulled-off the feat by training their system on hundreds of thousands of already published reviews.
Getting AI to review wine is relatively easy, they say, given that wine reviewers rely on a narrow, predictable vocabulary of standardized terms to rate wines.
Turns out, beer reviewers are similarly disciplined when it comes to evaluating beer.
So the researchers also designed the system to write beer reviews.
The reviews take the form of a few, pithy sentences, such as: “Pretty dark for a rosé, and full-bodied, with cherry, raspberry, vanilla and spice flavors. It’s dry with good acidity.”
The upshot: “The hope is that AI can benefit reviewers facing larger writing workloads and consumers who have to sort through so much content about products,” says Prasad Van, a member of the research team and a professor of business administration at Dartmouth.
In other AI-generated writing news:
*In-Depth Guide: Hyperwrite: Mike Kaput, a senior consultant at PR 20/20, a marketing agency, is all-in on AI writer HyperWrite.
Observes Kaput of the tool: “Artificial intelligence wrote an entire blog post for me — seriously, all of it.
“I gave AI a topic, and the machine did the rest.
“The result was a 100% factual, 100% original blog post that totaled about 650 words.”
“AI tools that generate language are nothing new.
“But the technology appears to have advanced to an unprecedented level.
“We might not be far away from machines writing in-depth long-form content.”
Click here for a detailed, step-by-step look at how HyperWrite works.
*AI Snares European Award for Automated Election Coverage: AI won a top award in the European Digital Publishing Awards for its automated coverage of elections for the Rheinische Post.
The AI tool churned-out dozens of hyper-local stories on election results — just minutes after the final vote counts were in.
Squiring the entire automated election process coverage was a single editor and manager, according to Retresco, the AI company behind the AI-generated writing tool.
*United Robots Automates Upper Midwest Ice Hockey Coverage: United Robots — a prolific provider and evangelist of automated journalism — has another conquest under its belt, this time in the Upper Midwest.
The AI company has helped Forum Communications go live with automated news coverage of two junior ice hockey leagues across Wisconsin, Minnesota, North Dakota and South Dakota.
AI affords us “unique opportunities to cover more teams, more players — in an efficient and creative manner,” says Neal Ronqist, general manager, Forum Communications.
*Press Releases: Ripe for AI Assistance: While writer Yakup Özkardes-Cheung insists humans still need to play an integral part in the writing of press releases, automation can certainly lighten the load.
Observes Özkardes-Cheung: “The good side of AI-generated press releases is that they are able to produce content at a faster rate than human writers.
“This is because they don’t need to sleep or take breaks.
“They also have the ability to write in different languages and tones, depending on what the client needs.
“This type of writing can be seen as an assistant — rather than a replacement — for human writers.
“It’s not meant to replace humans — but can instead help them do their job better and more efficiently.”
*Google’s AI Auto-Correct Editor — Maybe Too Correct?: Not everyone’s a fan of Google’s AI-powered auto-correct editor — which sometimes makes what some consider annoying suggestions, according to writer Katyanna Quach.
Observes Quach: “Words like ‘policemen’ could trigger the (Google) model into suggesting it be changed to something more neutral like ‘police officers.’ That’s understandable.
“But it can get a bit ridiculous. For example, it (Google Auto-Correct) proposed replacing the word ‘landlord’ with ‘property owner’ or ‘proprietor.'”
Moreover, when it comes to ensuring writing is fair and balanced, tools like Google Auto-Correct are facing an even tougher challenge, according to Quach.
Observes Quach: “For starters, there are apparently 21 definitions of fairness in academia.
“Fairness means different things to different groups of people.”
*AX Semantics’ Takes on Google’s Ruling: ‘AI Writing is Spam:’ AI-generated writing pioneer AX Semantics has a different take on a recent policy statement by Google that it sees AI writing as spam.
The company — in a post — bases its analysis on Google’s Webmaster Guidelines, which indicates that automated content is only a problem if it does not enhance the overall user value of content produced.
Even so, those who follow Google closely may see AX Semantics’ argument as hairsplitting.
Essentially: The reason why Google’s recent policy statement has turned so many heads is that it sounds so crystal clear.
Observes John Mueller, search advocate, Google: “For us, if you’re using machine learning tools to generate your content, it’s essentially the same as if you’re just shuffling words around, or looking up synonyms, or doing the translation tricks that people used to do.”
Bottom line: This is a controversy anyone working with AI-generated writing needs to follow closely.
*Google Snatches Bragging Rights to One of Biggest Auto-Writing Engines: In the never-ending horserace to build the most powerful supercomputer-powered auto-writer, Google currently has bragging rights.
It just released its biggest auto-writing engine yet: the Pathways Language Model (PaLM), boasting 540 billion parameters.
Observes Anna Wyndham, a writer for Slator: “PaLM is just a touch larger than Microsoft / NVIDIA’s Megatron-Turing NLG, almost double the size of DeepMind’s Gopher, and a whole lot bigger than Open AI’s GPT-3 (175 billion parameters).”
Since its unveiling in Summer 2020, GPT-3 has served as the underlying writing engine of more than 50 automated writing solutions.
Virtually all of these solutions run atop GPT-3, essentially serving as a software interface that users can leverage to operate the underlying GPT-3 auto-writing engine.
*EleutherAI’s Open Source Auto-Writing Engine: A Walkthrough: Champions of ‘free-for-the-people’ auto-writing engines will take cheer in this piece, which offers a step-by-step tutorial on how to use EleutherAI’s GPT-Neo — an auto-writing engine that is free to use.
Granted, the text is a tad technical.
But it’s perfect for someone dying to get a look into the inner workings of automated writing — and master the most powerful, free auto-writing engine that’s currently available.
*AI Big Picture: Top AI Conferences of 2022: Elizabeth Juran, a senior consultant at PR 20/20, offers an excellent overview of upcoming AI conferences in this piece.
The gatherings are perfect for writers and others looking to take a broader look at the current and projected impact of AI on business and society.
~World Summit AI Americas 2022 (Montreal and Virtual), May 4-5
~The AI Summit London 2022
May 27-29
~Deep Learning World (Las Vegas)
June 19-24
Share a Link: Please consider sharing a link to https://RobotWritersAI.com from your blog, social media post, publication or emails. More links leading to RobotWritersAI.com helps everyone interested in AI-generated writing.
–Joe Dysart is editor of RobotWritersAI.com and a tech journalist with 20+ years experience. His work has appeared in 150+ publications, including The New York Times and the Financial Times of London.