Skip to Content
Artificial intelligence

2017: The Year AI Floated into the Cloud

As more AI gets done on remote servers, Google, Amazon, and Microsoft are dueling to see who reigns supreme (and makes the most money).
December 29, 2017
Ms. Tech

Cloud computing is already a huge business, and competition is stiff. But this year, tech firms opened a new front in the battle to win users over in the cloud: the large-scale introduction of cloud-based AI.

For small and medium-size companies, building AI-capable systems at scale can be prohibitively expensive, largely because training algorithms takes a lot of computing power. Enter the likes of Amazon, Microsoft, and Google, each of which has vast stores of computing power and a big stake in the $40 billion cloud computing industry. For them, adding AI is simply a matter of keeping up with customers, who increasingly are looking for cost-effective ways of building machine learning into their software.

Amazon, with its AWS Amazon Cloud‎ service, has been leading the way. At the AWS conference in Las Vegas earlier this year, the company showed off Amazon Cloud 9, an integrated development environment (IDE) that plugs directly into its cloud platform. It also announced a host of new AI tools that can turn speech in audio files into time-stamped text, for example, as well as translate between seven languages and track people, activities, and objects in video.

Google lags behind Amazon and Microsoft in overall cloud services but is making a play for more market share with TensorFlow, open-source AI software that can build other machine-learning software. Since its launch, it’s become the AI platform of choice for many developers, and it underpins many new artificial-intelligence projects. The company has created its own chips, too, called Tensor Processing Units (TPUs), which are designed to efficiently process TensorFlow and cut down on energy needs.

Of course, Microsoft and Amazon aren’t giving up ground without a fight. In fact, they’re teaming up. The two launched an open-source deep-learning library called Gluon that works a lot like TensorFlow and is meant to make it as easy to build and train neural networks as it is to make an app. Microsoft is also trying out low-power chips to run its Azure cloud servers.

AI in the cloud is about more than just power plays by tech giants, though—it could also be behind the next leap forward in artificial intelligence. Rigetti Computing, a company in California, just used one of its prototype quantum chips to run a machine-learning algorithm on its cloud platform. The technology is so new that even experts are unsure what it is capable of. But one thing’s for sure: there will be a lot of learning done in the cloud in 2018.

Deep Dive

Artificial intelligence

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist

An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.

Unpacking the hype around OpenAI’s rumored new Q* model

If OpenAI's new model can solve grade-school math, it could pave the way for more powerful systems.

Generative AI deployment: Strategies for smooth scaling

Our global poll examines key decision points for putting AI to use in the enterprise.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.