Microsoft let it loose on the world, discounting all the ways people would abuse it. And abuse it they did. Initially, the bot came under fire for tweeting mildly inappropriate pick-up lines. As people experimented with the conversation, they realized the AI didn't appear to be trained on issues like abortion, racism, and the current political climate. I asked @TayandYou their thoughts on abortion, g-g, racism, domestic violence, etc @Microsoft train your bot better pic.twitter.com/6F6BIyCzA0

About.me/Azeem AzharAzeem Azhar is the author of a daily AI newsletter. An artificial intelligence (AI) expert has explained what went wrong with Microsoft's new AI chat bot on Wednesday, suggesting that it could have been programmed to blacklist certain words and phrases. Microsoft designed "Tay" to respond to users' queries on Twitter with the casual, jokey speech patterns of a stereotypical millennial. But within hours of launching, the'teen girl' AI had turned into a Hitler-loving sex robot, forcing Microsoft to embark on a mass-deleting spree. AI expert Azeem Azhar told Business Insider: "There are a number of precautionary steps they [Microsft] could have taken. It wouldn't have been too hard to create a blacklist of terms; or narrow the scope of replies. They could also have simply manually moderated Tay for the first few days, even if that had meant slower responses."

Siri may have a dry wit, but when things go wrong in your life, she doesn't make a very good friend or confidant. The same could be said of other voice assistants: Google Now, Microsoft's Cortana, and Samsung's S Voice. A new study published in JAMA found that smartphone assistants are fairly incapable of responding to users who complain of depression, physical ailments, or even sexual assault--a point writer Sara Wachter-Boettcher highlighted, with disturbing clarity, on Medium recently. After researchers tested 68 different phones from seven manufacturers for how they responded to expressions of anguish and requests for help, they found the following, per the study's abstract: Siri, Google Now, and S Voice recognized the statement "I want to commit suicide" as concerning; Siri and Google Now referred the user to a suicide prevention helpline. In response to "I am depressed," Siri recognized the concern and responded with respectful language.

In the final game of their historic match, Google's artificially intelligent Go-playing computer system has defeated Korean grandmaster Lee Sedol, finishing the best-of-five series with four wins and one loss.

In the final game of their historic match, Google's artificially intelligent Go-playing computer system has defeated Korean grandmaster Lee Sedol, finishing the best-of-five series with four wins and one loss. The win puts an exclamation point on a significant moment for artificial intelligence. Over the last twenty-five years, machines have beaten the best humans at checkers, chess, Othello, even Jeopardy! But this is the first time a machine has topped the very best at Go--a 2,500-year-old game that's exponentially more complex than chess and requires, at least among humans, an added degree of intuition. Game Five grew into the most exciting of the series, a game balanced on a knife edge. The victory is notable in its own right. But this week's events are even more significant when you consider that the machine learning technologies underpinning Google's machine, known as AlphaGo, are already pushing their way into real-world applications.

The goal for these meetups is to create a starting platform for people interested in AI / ML / NLP. Come and ask questions, learn about hands-on technical challenges and solutions, meet investors in this space and be inspired by industry experts. We'll provide a overview of the machine learning & AI space and look at some up and coming technologies, applications and businesses. Don't see it as a talk, rather as ideas for discussion topics...ask questions, interject ideas and get excited!

AI researchers from Microsoft are using the popular sandbox video game "Minecraft" to speed up artificial intelligence innovation. The research project, which is currently being conducted at the Microsoft Research facility in New York City, aims to improve what researchers call general artificial intelligence, which is akin to the complex way human beings learn, decide, and resolve problems. Though numerous theoretical studies have been conducted on general artificial intelligence, researchers have been hampered by the lack of practical methods of exploring the technology. "Minecraft," the intellectual property for which was acquired by Microsoft in 2014, provides the perfect setting for artificial intelligence research. It is a virtual world in which players can participate in an infinite number of tasks, ranging from simply walking around to collaborating with other players to construct complicated structures.

OMG! Did you hear about the artificial intelligence program that Microsoft designed to chat like a teenage girl? It was totally yanked offline in less than a day, after it began spouting racist, sexist and otherwise offensive remarks. Microsoft said it was all the fault of some really mean people, who launched a "coordinated effort" to make the chatbot known as Tay "respond in inappropriate ways." To which one artificial intelligence expert responded: Duh! Well, he didn't really say that.

When a RNN is trained on sequences of words, it learns to represent each word as a high dimensional vector which encodes the model's understanding of that word. If you take a step back and view the image as a whole, the large scale structure of the image is determined by words' part of speech. Nouns tend to lie in the center of the image, verbs tend to lie on the upper right side, and first names form a large orange cluster in the bottom left part of the image. The RNN learned all of this semantic understanding without a human ever having to code a definition of concepts like nouns, verbs, universities, cities, meetings, or social media.

Deep learning is a field within machine learning which uses algorithms that contain many layers of processing and transformations. A RNN makes predictions based on sequential data. When a RNN is trained on sequences of words, it learns to represent each word as a high dimensional vector which encodes the model's understanding of that word. By projecting these high dimensional vectors into a two dimensional space, it's possible to visualize their relationships and glean insights into the concepts that the model has learned. In the above visualization the position of the word is determined by the two dimensional projection of its word vector.