In February, I took a job designing the personality of a chatbot called Kai. I ghostwrite the lines it says, and I have thought, while testing it, that talking to myself has rarely been so unpredictable. Kai, which was conceived by my employer, Kasisto, to help customers with online banking, works over text message, Slack, and especially Facebook Messenger, where more than thirty-four thousand other chatbots have joined it since April, when Facebook opened the platform to developers. Many of these bots possess no personality. The ones created by CNN and the Wall Street Journal, for instance, greet first-time users with "we," as if the whole newsroom were on the other side of the screen, and run keyword searches rather than engaging in conversation.

Kaggle has become the premier Data Science competition where the best and the brightest turn out in droves – Kaggle has more than 400,000 users – to try and claim the glory. With so many Data Scientists vying to win each competition (around 100,000 entries/month), prospective entrants can use all the tips they can get. And who better than Kaggle CEO and Founder, Anthony Goldbloom, to dish out that advice? We caught up with him at Extract SF 2015 in October to pick his brain about how best to approach a Kaggle competition. According to Anthony, in the history of Kaggle competitions, there are only two Machine Learning approaches that win competitions: Handcrafted & Neural Networks.

We've heard that it is prohibitively expensive for startups and academics to train machine learning models, and this is due to the rental or purchase costs of hardware. The results from one recent Google paper were estimated to cost $13k to emulate. That's just to reproduce the final model, not to emulate the whole experimentation and hyperparameter optimisation caboodle. Equally, there are intelligence tasks (training, inference, or prediction) that would ideally happen on the cellphone or remote sensor but are too compute constrained locally, so currently rely on uploading data to the cloud for processing. Machine intelligence is the future of computing, so what needs to happen at a hardware level to make it faster and more energy- and cost-efficient? We talked to Simon Knowles, CTO of Graphcore, about hardware acceleration of machine intelligence.

I've always wanted to like The Elder Scrolls V: Skyrim more than I ever have in practice---a problem I had with Oblivion as well, and one that I have with lots of RPGs (as well as MMOs like The Elder Scrolls Online.) Skyrim is riddled with problems, of course, from bugs and glitches to uneven acting and quest design. But it's always been the lackluster combat that left me cold. And I hate that, because this is exactly the kind of game I dreamed about as a kid: Sprawling, epic, filled with secrets and quests and magical items, dark secrets and shadowy organizations. And yet every time I play the game I quickly get bored, or find myself switching to something else with more engaging mechanics like Dark Souls.

December 27, 2016 --With excitement swirling around the possibility of a ninth planet, a rebound in the global tiger population for the first time in a century, and the DNA sequenced in space for the first time, 2016 has been a year full of scientific wonder. But as the year comes to a close, there are some breakthroughs particularly worth highlighting. In February, a century after Albert Einstein predicted their existence, an international team of researchers confirmed that they had actually detected a ripple in the fabric of spacetime for the first time. The detection of gravitational waves came across as a "chirp" across the detectors that make up the Laser Interferometer Gravitational-wave Observatory (LIGO), but the researchers say it was the result of two large celestial bodies, possibly black holes, colliding some 1.3 billion years ago. Then, in June, the scientists announced that the cosmos had chirped again.

December 27, 2016 --With excitement swirling around the possibility of a ninth planet, a rebound in the global tiger population for the first time in a century, and the DNA sequenced in space for the first time, 2016 has been a year full of scientific wonder. But as the year comes to a close, there are some breakthroughs particularly worth highlighting. In February, a century after Albert Einstein predicted their existence, an international team of researchers confirmed that they had actually detected a ripple in the fabric of spacetime for the first time. The detection of gravitational waves came across as a "chirp" across the detectors that make up the Laser Interferometer Gravitational-wave Observatory (LIGO), but the researchers say it was the result of two large celestial bodies, possibly black holes, colliding some 1.3 billion years ago. Then, in June, the scientists announced that the cosmos had chirped again.

GE was still essentially a startup when its managers hired young MIT chemistry professor Willis Whitney to open the company's first laboratory in 1900. Unlike Thomas Edison's Menlo Park "invention factory" in New Jersey, the place was a modest affair. It was located in a wooden barn behind the house of GE's resident polymath and engineering wizard Charles Steinmetz in Schenectady, New York, where GE co-founder Edison moved in the 1880s. The lab -- dedicated to "fundamental research" -- went up in flames the next spring, but Whitney's career and momentum continued. He rebuilt the lab and brought in researchers such as Nobel Prize-winning chemist Irving Langmuir.

How can we encourage more young women to become involved in computer science? I think that there have been great strides in encouraging women and minorities to consider studying computer science. I see increased interest and participation among women and minorities in my community and it is encouraging, however, I also recognize that I live in a pretty unique place. My sister often tells me that her town just doesn't offer the same programs. To combat this type of challenge, I think we need to continue to normalize the idea of females in Computer Science and also expand the programs outside of big cities and academic towns.

Owing to the data deluge and the Cambrian explosion of machine learning techniques over the past decade, one might have expected the transformation of marketing strategy into a predominantly quantitative discipline by now. The fact that it hasn't happened yet, and the observation that marketing is still influenced by a lot of qualitative inputs can be ascribed to two reasons, in my opinion. The first and principal reason continues to be institutional inertia. Second, there is a significant communication and knowledge gap between data scientists and marketers, owing to their relative lack of familiarity with the other side's perspectives and paradigms. The successful marketer of the next decade is someone who is conversant with management theories of Kotler[1] as well as machine learning advances by Hinton[2]/LeCun[3]/ Ng[4].

Hollywood actor Morgan Freeman has provided the voice for an AI assistant created by Facebook's Mark Zuckerberg. Mr Zuckerberg said he asked the actor, who was chosen by the public, after an awards ceremony earlier this month. The Facebook co-founder coded the AI assistant - called Jarvis, after the butler in Iron Man - for his home. If he decides to release it to the public, people would relate differently to a famous voice than more robotic sounding assistants, tech experts said. Mr Zuckerberg asked his Facebook followers to pick the voice after building artificial intelligence to help him around the house.