No, we’re not talking about sentient killer robots, and yes, AI is more than self-driving smart cars and chat bots. Natural language processing, deep learning, and machine learning are changing the way businesses are interacting with customers and how platforms are serving users.
AI is certainly being touted as the next big thing, but many companies have already hit the ground running—and only gained speed since getting there—by building machine learning into the core functions of their business.
Here are 5 companies nailing this machine learning thing, and how they’re doing it:
Amazon
As one of the largest retailers in the world, Amazon also boasts one of the largest AI platforms in the world. Whether you’re shopping for paper towels or functioning within their AWS platform, chances are you’re engaging with several machine-learning programs.
On the retail side, everything from product recommendations to supply chain, forecasting, and capacity planning runs on machine-learning, while programs like Macie and Glue that scan for sensitive data breaches and perform data cleansing, respectively. Of course, let’s not forget Alexa, Prime Air, and Amazon Go, which all function through AI algorithms, while rumors of an AI fashion designer are feeding the Amazon AI flame.
Netflix
Netflix’s AI isn’t just suggesting what show you watch after Stranger Things (though it does that too, and brings in an estimated $1 billion a year alone by delivering personalized experiences to every user), but earlier this year, the online streaming giant announced an AI algorithm called Dynamic Optimizer to analyze each and every frame of video in each of it’s roughly 13,000 titles it streams and compresses it without sacrificing image quality. According to
Ioannia Katsavounidis, Sr. Research Scientist at Netflix, “We can now optimize scene by scene with an almost infinite matrix of possibilities.” This reduces the amount of data it takes to stream a video and optimizes for lower-bandwidth viewing on mobile devices. In short, it keeps your videos from buffering.
It’s no surprise that Google, or its parent company Alphabet, is at the forefront of innovation. Google was one of the pioneers of machine learning with suggested searches and ever-evolving search ranking algorithms. Google’s Machine Intelligence efforts have focused on deep learning, which involves multiple layers of neural networks—built to simulate human thought processes—that allow Google’s technology to process data more thoroughly.
Remember image recognition? Now Google’s taking it a step further with Image Enhancement, which fills in missing details in images with high-resolution information. In short, it can fill in details in pixelated images, making “zoom and enhance” a reality.
And Google’s machines have been using image recognition within artificial neural networks to dream for the past two years. Google trained the networks to extract important information, ignore irrelevant information, and generate images in “dreams.” It’s pretty complex, but you can learn more about it here.
Salesforce
Salesforce has named it’s AI Einstein, and launched it at their 2016 Dreamforce conference. It stretches across their cloud services, from sales to marketing, services, community, and apps, powering nearly every part of their product with machine-learning algorithms.
Why Einstein? To explain, they cite Albert Einstein himself when he said, “the definition of genius it taking the complex and making it simple.” And for the most part, that’s exactly what Einstein does. It enables users to create smarter, personalized, and more predictive customer experiences. Einstein lives within a set of services within their Customer Success Platform and provides customized models to every customer, then adjusts with every customer interaction and additional piece of data it receives. Einstein is designed to automatically discover relevant insights, predict customer behavior, recommend next actions to users, and automate tasks, like logging customer data, capturing sales activity, scoring leads, delivering content, and messaging customers when they’re most likely to engage.
IBM
You’ve undoubtedly heard of IBM’s Watson by now; the computer that won Jeopardy back in 2011. But Watson’s been following self-learning behavior models and has done everything from diagnosing certain types of cancers more effectively than oncologists, writing songs, and producing movie trailers.
Watson was designed, however, to consume data at a drastically faster rate than a human can, to learn from it, and augment human capabilities. In the case of cancer treatments, Watson can read half a million medical research papers in 15 seconds and was trained at Memorial Sloan Kettering in New York to be able to suggest diagnoses and treatments to doctors.
Simply put, IBM’s flagship AI—as with all machine learning programs—can do in nano-seconds what would take a human years—or decades—to do, and is transforming and optimizing business in the process.
Here at Domo, machine-learning is fully integrated into the Domo platform. Domo’s AI delivers deeper more advanced insights, powers business alerts, and drives intelligent recommendations to users, all to help businesses make
better decisions with the data they already have. Want to see it for yourself? Get Domo for free and see exactly how machine learning, paired with real data insights, changes everything.