Strong Amazon Web Services (AWS) results helped Amazon report stronger results than a year ago in the fourth quarter (ended Dec. 31), during which the company unveiled multiple new innovative AWS offerings that included five artificial intelligence (AI) services.
AWS revenue grew to $10 billion from $7.4 billion, while AWS operating profit increased to $2.6 billion from $2.2 billion. Total Amazon revenue jumped 21% to $87.4 billion, while profit grew to $3.3 billion ($6.47 a share) from $3 billion ($6.04 a share).
On an earnings call Jan. 30, Amazon CFO Brian Olsavsky chalked up the stronger AWS performance to a “combination of increased sales support, more and better products that hit customers’ needs and also [its] geographic expansion.”
It’s “kind of the culmination of a lot of work on adding new products and features, adding to our sales and marketing teams and having better penetration in enterprise customers and hitting a lot of very different industries,” he told analysts.
Meanwhile, “we continue to expand” AWS to more regions, he said, noting it’s “now in 69 availability zones across 22 geographic regions.”
The five new AWS AI services the company introduced were “designed to put machine learning in the hands of more developers — with no machine learning experience required,” the company said in its earnings news release.
The first new AI service, Amazon Kendra, “reinvents enterprise search by using natural language processing and other machine learning techniques to unite multiple data silos inside an enterprise and consistently provide high-quality results to common queries instead of a random list of links in response to keyword queries,” the company said.
The other new AI services are: Amazon CodeGuru, which Amazon said helps software developers automate code reviews and identify an application’s most expensive lines of code; Amazon Fraud Detector, which it said “helps businesses identify online identity and payment fraud in real time, based on the same technology developed for Amazon.com”; Amazon Transcribe Medical, which offers healthcare providers “highly accurate, real-time speech-to-text transcription so they can focus on patient care”; and Amazon Augmented Artificial Intelligence (A2I), which helps developers validate machine learning predictions through human confirmation.
AWS also announced a set of machine learning-powered analytics capabilities for Amazon Connect called Contact Lens, which, the company said, “make it easier for businesses to identify customer issues and trends, search call and chat transcripts, and improve agent performance.” Amazon Connect offers customers a “fully-managed cloud contact center service,” the firm said, adding: “Now with Contact Lens, Amazon Connect customers have the ability to understand the sentiment, trends, and compliance of their own customer conversations to improve the experience and identify crucial feedback, with no machine learning experience required.”
AWS, meanwhile, announced major new analytics capabilities in its Redshift data warehouse product that, it said, “provide an order of magnitude better query performance, deliver greater flexibility, and help customers embrace data at scale.”
Also new are three new AWS services and capabilities that Amazon said “make it easier for customers to build and operate securely.” The first, Amazon Detective, “analyzes trillions of data points, using machine learning, statistical analysis, and graph theory to make it easier to visualize and conduct faster and more efficient security investigations,” Amazon said.
AWS IAM Access Analyzer, meanwhile, “makes it simple for security teams and administrators to audit resource policies for unintended access by analyzing hundreds or even thousands of policies across a customer’s environment in seconds, and delivering detailed findings about resources that are accessible from outside the account,” Amazon said.
Last, AWS Nitro Enclaves is a new Amazon Elastic Compute Cloud 2 capability that it said “makes it easy for customers in healthcare, financial services, energy, media and entertainment, and other data-intensive industries to process highly sensitive data, like personally identifiable information and intellectual property on their compute instances, particularly from internal threats within their own accounts.”
AWS also announced six new capabilities for its Amazon SageMaker managed service that it noted “removes the heavy lifting from each step of the machine learning process.” Amazon SageMaker Studio is a fully integrated development environment for machine learning that, it said, “makes it easier for developers to build, debug, train, deploy, monitor, and operate custom machine learning models.”
Amazon SageMaker Notebooks, meanwhile, “allows developers to spin up elastic machine learning notebooks in seconds, and automates the process of sharing notebooks with a single-click,” the company said.
Amazon SageMaker Experiments helps developers visualize and compare machine learning model iterations, training parameters and outcomes, while Amazon SageMaker Autopilot allows developers to submit simple data in CSV files and have machine learning models automatically generated, with “full visibility to how the models are created so developers can evolve them over time,” Amazon said.
Also, Amazon SageMaker Debugger “provides real-time monitoring for machine learning models to improve predictive accuracy, reduce training times, and facilitate Adrift to discover when the performance of a model running in production begins to deviate from the original trained model,” Amazon said.
Amazon also announced there are now “hundreds of millions” of Alexa-enabled devices “in customers’ hands and customers interact with Alexa billions of times each week.”
The company’s Fire TV, meanwhile, now has more than 40 million active users globally, it said.