Hypercloud Guys Make Big AI Push
AI is emerging as a competitive arena for hypercloud providers, and it seems like all those guys had a big surge Tuesday. Microsoft, Google and Amazon all introduced AI and machine learning technology -- as well as services in the related field of analytics -- with Tesla Motors' Elon Musk getting into the act.
Microsoft Corp. (Nasdaq: MSFT) Azure on Tuesday announced a partnership with OpenAI, a nonprofit AI research organization backed by Musk, LinkedIn co-founder Reid Hoffman and controversial tech investor Peter Thiel. OpenAI will use Azure as its primary cloud platform.
Azure announced a new class of virtual machines, called the Azure N-Series Virtual Machine, using NVIDIA GPUS for the most intense compute workloads, including deep learning, simulations, rendering and training neural networks. OpenAI will be one of the first organizations to use the new machines.
Microsoft also launched Azure Bot Services in preview, to allow developers to build, connect, deploy and manage intelligent bots that interact naturally on apps, websites, texting, Slack, Facebook Messenger, Skype and more.
Meanwhile, Google (Nasdaq: GOOG) launched a new Google Cloud Machine Learning group focused on delivering cloud-based machine-learning solutions to businesses, headed up by researchers Fei-Fei Li and Jia Li.
Google also launched its Cloud Jobs API, which is designed to help businesses match relevant jobs to candidates and is intended for job boards, careers sites and applicant tracking systems, with early adopters including Jibe, Dice and Careerbuilder.
The Google Cloud Jobs API is a competitive thrust at LinkedIn, which Microsoft is in the process of buying.
Like Microsoft Azure, Google also introduced support for GPUs on its cloud platform Tuesday, to provide the processing horsepower for machine-learning applications.
And Google is using custom ASICs for machine learning to reduce the price for its Cloud Vision API by about 80%, as well as improving image recognition -- for example, identifying logos, labels, landmarks and other objects in images.
Google introduced a premium version of its Translation API for improved quality, and support for up to eight languages, for users requiring precise, long-form translation services, such as livestream translations, high email volumes, detailed articles and documents.
And Amazon Web Services Inc. announced Tuesday that QuickSight is available to all customers, with prices starting at $9 per user per month. The service is designed to enable all employees within a company, regardless of their technical skill, to build visualizations, perform ad-hoc analysis and get business insights from their data.
AI requires specialized processing power and a lot of talent -- difficult for even multi-billion-dollar enterprises to amass but within the reach of hypercloud providers. And those same hypercloud providers can use standardized interfaces and languages to offer AI services on a broad scale, for mutual profit.
- Amazon Unlimited, Alexa, AI & How to Cut Off Google at Its Knees
- Say What Now? Samsung Buys Into AI Assistants With Viv
- Google's New Dream: Pixel, VR & an AI Assistant With Smarts
- Microsoft Bets Big on AI Future
- Microsoft Launches AI & Research Group
- AI Paving the Way for 5G, IoT
- Microsoft & LinkedIn: Marriage Made in the Cloud
— Mitch Wagner, , Editor, Light Reading Enterprise Cloud