For the first time, Internet video pioneer YouTube has had to take into account that its dominance is being challenged, most notably by Facebook, Snapchat and Amazon. Now, YouTube has more than a billion users, an app audience of 18-to-49-year-olds that dwarfs that of any U.S. cable network, and an average mobile viewing session more than 40 minutes long. To keep ahead of the competition, YouTube has diversified, with apps devoted to specific niche audiences: YouTube Music, YouTube Kids and YouTube Gaming. Continue reading YouTube Faces Rivals, Diversifies to Capture Niche Audiences

Zendesk is introducing Automatic Answers, a service that answers a business’s emails from customers, without human intervention. But it’s far more sophisticated than today’s well-known auto-responders, having been built on a machine-learning platform that in-house data scientists and engineers have been developing for some time. The platform is also used for another Zendesk service, Satisfaction Prediction, which helps determine the customer experience. Automatic Answers will be rolled out later in the year. Continue reading Zendesk Customer Service Product Built on Machine Learning

Next month, Snapchat will introduce Memories, a new feature designed to let users save favorite photos and videos, re-edit them by adding filters and text and then re-share the latest versions. Included is a search function that lets the user search Memories by date, location or keywords. Prior to Memories, a Snapchat user could save a photo or video by downloading it to her smartphone, but couldn’t view it in the app or share it. Memories still does not allow a Snapchat user to save a photo/video sent by someone else. Continue reading Snapchat to Debut Memories to Save, Tweak, Re-Share Snaps

Anki, the company that introduced Anki Drive, artificial intelligence-powered racecars, will ship its second product in October: a robot dubbed Cozmo, the size of a coffee mug that costs $180. By combining artificial intelligence, computer-vision science, advanced robotics, character development and machine-learning algorithms, Cozmo is a toy that is intended to be much like a real-world Wall-E or R2-D2, says chief executive Boris Sofman. The company is also offering SDKs for Cozmo’s components. Continue reading Cozmo: Anki to Launch Robot Powered by Machine Learning

Twitter announced it is acquiring London-based artificial intelligence startup Magic Pony Technology to help provide a professional polish to tweeted live videos. The social giant reportedly paid about $150 million to purchase Magic Pony. Twitter CEO Jack Dorsey said he was buying the company “so Twitter can continue to be the best place to see what’s happening and why it matters, first.” Twitter has been emphasizing video in recent months, and machine learning is “increasingly at the core of everything we build,” said Dorsey. In addition to using machine learning, “Magic Pony’s technology uses artificial intelligence for visual effects,” notes Bloomberg. “It can be used to clean up pixelated images or create new images” and “to improve video streaming.” Continue reading Twitter Eyes Machine Learning with Acquisition of Magic Pony

Microsoft just paid $26.2 billion to buy LinkedIn and its treasure trove of information about the business social network’s 105.5 million monthly active users. That enormous quantity of data will drive machine learning to create and evolve products as well as anticipate customer needs. Microsoft isn’t alone in going after large repositories of data for that very reason. As Microsoft, Apple and Alphabet exploit the possibilities of machine learning, they all are competing for the information necessary to find actionable patterns. Continue reading Microsoft to Mine LinkedIn Data, Advance Machine Learning

At its annual shareholders meeting, Alphabet chairman Eric Schmidt reiterated the company’s plan to wirelessly connect homes to high-speed Internet. The technology, viable now due to improvements in computer chips and more accurate wireless signal targeting, can deliver a 1 gigabit per second connection, equivalent to Google Fiber’s capabilities. Schmidt also delved into the company’s plans to advance search features, explore 3D printing for building construction, meatless meat, autonomous cars and other sectors. Continue reading Alphabet Update on Wireless Internet, Search, Education, Cars

Slack has plans to evolve its workplace app to become the center of the office’s digital environment, connecting a wide variety of third-party business apps, with the ultimate goal of becoming the artificial intelligence-powered portal to the entire workplace. Head of search, learning and intelligence Noah Weiss described those plans for Slack, currently used by three million people every day in the work environment, and added that the company is also evolving its voice capabilities to become more sophisticated. Continue reading Slack Strives to Be Artificial Intelligence-Powered Office Portal

Amazon is testing an as-of-yet unannounced new cloud service that will let businesses run a wider range of artificial intelligence software on its computers, say people close to the situation. This move puts Amazon, which launched Amazon Web Services in a limited offering in this area last year, in closer competition with Google, Microsoft and IBM, which have already launched various cloud services. The new service will help development of pattern recognition, speech transcription and other robust applications. Continue reading Amazon Creating New Cloud Services for Artificial Intelligence

During the Moogfest music and technology fest in North Carolina, Google Brain researcher Douglas Eck outlined a new artificial intelligence research project at Google called Magenta. The group, expected to publicly launch next month, plans to use the company’s machine learning engine TensorFlow to explore new ways that computers and AI systems could be trained to create original art and media such as music or video. The initiative should prove challenging; so far, the most advanced AI systems have struggled to replicate styles of existing artists. Continue reading Google to Explore Using AI Systems to Produce Art and Music

Storage is quickly becoming a crucial area of interest, as data grows exponentially. The world’s one billion cell phones already generate 18 exabytes (1 billion gigabytes) of data monthly and, as the Internet of Things places sensors in everything from automobiles to homes, the data output is estimated to soar into the zettabytes. Artificial intelligence and machine learning are a focus to find ways to help manage huge amounts of data. New ways to store data are imperative, and some practical advances are being made. Continue reading As Data Surges to Zettabytes, Need for Better Storage Urgent

Google has just built its own chip as part of its efforts to speed up artificial intelligence developments. The company revealed that this is just the first of many chips it plans to develop and build. At the same time, an increasing number of businesses are migrating to the cloud, lessening the need for servers that rely on chips to function. That’s led some to believe that Google and other Internet titans that follow its lead will impact the future of the chip industry, particularly such stalwarts as Intel and Nvidia. Continue reading Google Develops its Own Chip to Speed Up Machine Learning

Google is focusing on advances in artificial intelligence as a competitive advantage. Its new Google assistant is positioned as the company’s latest iteration of search, and Google is promoting a variety of products based on machine learning that face competing products. Google Home will compete with Amazon Echo, while messaging app Allo will take on Facebook’s Messenger and WhatsApp, and Snapchat. With regard to virtual assistants, Google is going against versions from Apple, Amazon, Facebook and Microsoft. Continue reading Google Launches its AI Ecosystem with Allo and Google Home

Google is open-sourcing SyntaxNet, a neural network framework that provides a foundation for Natural Language Understanding (NLU), and Parsey McParseface, a computer program that helps machines understand written English. Offering the code for free lets anyone develop, modify and distribute it, furthering natural language and potentially making Google’s code the standard. Earlier, Google open-sourced its machine-learning code TensorFlow (which SyntaxNet runs on top of); other companies that have similarly open-sourced code include Amazon and Facebook.

Continue reading Google Open Sources Language Tools for Virtual Assistants

The security research community has warned that the Internet of Things, including home security systems, is hackable. Researchers at the University of Michigan and Microsoft have published an in-depth security analysis of Samsung’s SmartThings platform that allows control of home appliances via PC or smartphone. They were able to hack the system, setting off a smoke alarm and opening a digital lock with a “backdoor” PIN. Their findings will be presented at the IEEE Symposium on Security and Privacy later this month. Continue reading Researchers Warn That Smart Home Technology is Hackable