Reddit is known for many things: lively communities, a dedicated user base, cum boxes, incest. Now, the Elon Musk-and-Peter Thiel-backed nonprofit OpenAI wants to use Reddit's vast array of content as a guide for its new machine learning programs. MIT Technology Review reports that OpenAI has partnered with NVIDIA to use the latter company's new DGX-1 supercomputer to train its deep learning systems both more rapidly and with more data. One way they're going about that, apparently, is by using Reddit, so cross your fingers that the robots don't start spouting abuse and garbage! "One very easy way of always getting our models to work better is to just scale the amount of compute," OpenAI research scientist Andrej Karpathy said in a press release. "So right now, if we're training on, say, a month of conversations on Reddit, we can, instead, train on entire years of conversations of people talking to each other on all of Reddit."

Come join us as we reimagine the world of Sherlock Holmes! We're pleased to partner with the Columbia University Digital Storytelling Lab, IBM and the Made in NY Media Center to present a day long Sherlock Hack that mixes storytelling, the Internet of Things & Artificial Intelligence. Together we'll explore how to harness various Watson APIs and the Bluemix platform to create a number of enchanted objects. This hack is the first in a series of events that will help to inform a special presentation during the New York Film Festival on Oct 1st & 2nd. Work done within the hacks and meetups will become part of a massive connected crime scene that transforms Lincoln Center into a collaborative story and play space.

Elon Musk has a well-documented fear of evil artificial intelligence, so it's no surprise filmmaker Werner Herzog sought him out for Lo and Behold: Reveries of the Connected World. Instead, the tech mogul is worried about A.I. that does whatever it takes to accomplish its task. "The biggest risk is not that A.I. will develop a will of its own," Musk says in a short clip of the new film obtained by Fortune. "But rather it will follow the will of its utility function or optimization function." Herzog, the man who made Grizzly Man, is more attuned to nature than future technologies.

It's entirely possible Werner Herzog could find philosophical wonders and dilemmas making a documentary about your shoe collection, but until then we'll have to settle for this prolific filmmaker's abiding interest in the vastness of humankind's dreams, desires and actions. His latest nonfiction foray is about no less than what's changed life as we know it these last few decades: that coursing, unseen river called the Internet. And though the German auteur claims to be a technophobe, "Lo and Behold: Reveries of the Connected World" is just the kind of percolating, wry probe we need into this fast-moving, digitally monopolizing age. Herzog ("Grizzly Man," "Cave of Forgotten Dreams"), a natural alarmist at the same time he thrives on humanity's boldness and invention, isn't the type of explorer who starts with a pre-arranged idea. What animates "Lo and Behold" is his questioning spirit regarding the Web's journey from host-to-host communications tool devised in a UCLA lab in 1969 to the scarily inter-reliant nervous system of today. But with Herzog's initially chronological approach, there's plenty of humor in the recollections of gray-haired pioneers like Leonard Kleinrock, who mentions how the first machine-to-machine message from the refrigerator-sized computer at UCLA resulted in -- what else? -- a crash.

Before we get to the features, let's talk about that design. The egg-like form, and bold colors are striking for sure -- if Ikea did wireless audio, it might look like this. The speaker uses four 40mm drivers to pump out sounds in 360-degrees, and there's a shiny glass touch-sensitive button on top so you can control your music with gestures. Physical controls are old news though, right? Whyd thinks so, and has given the speaker Amazon Echo-like voice control.

Like lots of people who've cut cable TV, I couldn't have done it without Netflix. For a fraction of a typical cable bill, Netflix provides more TV shows, movies, and comedy specials than I have time to watch. And while I sometimes turn to other streaming services (and my over-the-air antenna) for specific content, most evenings I'll just thumb through the Netflix app until I find something that grabs my attention. Apart from not supporting 4K resolution or a voice-control remote control, Roku's latest streaming stick delivers the best price/performance ratio of any media-streaming device on the market. This is precisely how Netflix wants people to behave.

With this month finally seeing the release of No Man's Sky, in which players can investigate and explore a huge universe full of diverse planets, I couldn't help but be reminded that although outer space has always held a fascination for many, there's another largely unexplored area much closer to home – the ocean. Although water makes up about 70% of Earth's surface, it is estimated that humans have only explored around 5% of the seabed. Aquatic legends such as Atlantis, the Bermuda triangle, and the disappearance of the USS Scorpion in 1968 have captured the imaginations of many a budding oceanographer, and as such the depths of the sea are an unknown region still ripe for mythologizing – which is exactly what Abzu does, in the first release from Giant Squid software. Directed by Matt Nava, who worked as art director on Journey and Flower for thatgamecompany, the style is instantly reminiscent of his previous work, with players taking control of a stylised'Diver' character. Dropped into the ocean with little explanation, the first thing you notice is how - no matter how long you spend underwater - there's no air gauge to run out.

I joined Spotify in 2008 to focus on machine learning and music recommendations. It's easy to forget, but Spotify's key differentiator back then was the low-latency playback. People would say that it felt like they had the music on their own hard drive. In 2009 after a crazy amount of negotiation the music labels agreed to try it out as an experiment. Music distribution is a trivial problem now.

Earlier this year, Adobe took the wraps off its new Adobe Marketing Cloud, touting new data science capabilities like Adobe Analytics' Segment IQ, which uses machine learning to help marketers gain deep insight into audience segments. On Wednesday, Adobe advanced Segment IQ another step with the release of Segment Comparison for Analysis Workspace. Segment Comparison for Analysis Workspace is the first in what Adobe promises will be a series of audience analysis and discovery tools within Segment IQ. It uses machine learning techniques to perform automated analysis on every metric and dimension to which you have access. Nate Smith, senior product marketing manager, Adobe Analytics, says this allows Segment Comparison to uncover the key characteristics of the audience segments that are driving your company's KPIs.

Remember the horror movie The Fly, in which a brilliant scientist comes up with an groundbreaking technology -- only for his good intentions to head south when he accidentally gets genetically spliced with a housefly, turning him into a hideous monster? That's kind of the vibe we get hearing about a project that aims to teach an AI to hold a polite conversation by feeding it message threads from Reddit. Nonetheless, that's the working theory held by researchers at OpenAI, the nonprofit Artificial Intelligence company backed by Silicon Valley heavy-hitters like Elon Musk, which lists its mission as "to build safe AI, and ensure AI's benefits are as widely and evenly distributed as possible." All kidding aside, the impressive-sounding project focuses on creating a deep learning neural network able to build up a "probabilistic understanding" of conversation, and maybe even construct enough of a language model to be able to communicate itself. To help with this, OpenAI is employing a DGX-1 supercomputer developed by Nvidia.