Just as fintech is transforming the banking world, "insurtech" has set its sights on the insurance industry. There are firms like Trov, which provides insurance on-demand and enables consumers to catalogue insured belongings. Buzzmove uses home removals data to create a new distribution model for insurance – people are more likely to renew or buy insurance when moving house – and helps customers track the value of their assets. FitSense helps employers leverage data from wearable tech, while massUp capitalises on open APIs to connect insurers to retailers. As a consumer, you can cover your gadget at the point of sale – ideal if you know it might not be covered by your household policy. Start listing the players currently in the market, though, and while the added visibility and speed brought by innovative technology is evidently beneficial, "'disruption' can be a misleading word", says Jonathan Howe, UK insurance lead at PwC. "We take that to mean the whole industry has been thrown in the air and we've seen major new players emerge.

I visited my local health-insurance office a few months ago. After entering the building, I was welcomed into a long and dark corridor, full of nervous people carrying bloated folders. The atmosphere was gloomy, and it was obvious that no one wanted to be there. After about 30 minutes I realized why: During that time, the line had barely moved, and it took me the better part of the day to reach a clerk. As a result, I was late for two other errands I had planned.

Thanks to many impressive and rapid advances over the past few years, Artificial Intelligence has come into the spotlight – again. This could just be one of many false starts that this field has seen over the many years of its existence, or yet another Silicon Valley fad designed that startups are using to quickly raise money and gain market share, but there are many good reasons to believe much of the hype. Artificial Intelligence is probably here for good, and much of the technology and business in the upcoming years will be focused on the best ways of leveraging it. Judging by its title, this short ebook seems to provide a good explanation of what Artificial Intelligence is. Sadly, this is not the case.

John Searle, the Slusser Professor of Philosophy at the University of California, Berkeley, talked at Google about the philosophy of the mind and the potential for consciousness in artificial intelligence. Searle is widely known for his contributions to various fields of philosophy. First, Searle defined some terms he will be using for the rest of his talk and some types of phenomena. He stated that consciousness is an ontologically subjective domain, but that doesn't prevent epistemologically objective accounts of it. He also claimed that computers have no awareness of states of systems.

Tiro Partners is representing a major high street and online retailer in central London who are looking to build out a Recommender based system and a Forecasting System. The successful engineer must have a deep understanding of the algorithms used and why they are apprpriate. He/she must also be familiar with Workflow definition, Model Taining and Data Regularisation. Communication and mentoring skills are also essential; will be expected to explain what is being done and interact with a team of very talented but inexperienced engineers in the Machine Learning field. The initial work will focus on building the recommendations engine that will be use in the eCommerce site.

I have to admit that when "Big Data" starting getting a lot of traction, I was not a fan of the phrase as I often work with organisations who do not think of themselves in terms of having a lot of data. We can talk all we want about the multiple dimensions of big data, commonly referred do as the "V's" (volume, variety, velocity and veracity) but what sticks is the term "Big Data" and that conjures up images for people, after all marketing is powerful tool! Even today, I often have discussions regarding how a specific customer can take advantage of the new capabilities introduced in the era of big data, regardless of the size of the data stored. There are many different genres of analytics available that can be utilised to extend your analytic capabilities. Three of the most common include graph analysis, path analysis and text analysis.

Google acquired Kifi, an app for collecting links from across the Internet, which teams could then collaborate on. No financial terms of the deal were disclosed. The Kifi service and data will not become part of Google. The service will remain fully functional for existing users for a few more weeks, before its shut down. The company is no longer accepting new registrations. The team at Kifi will be joining the Spaces team at Google.

Editor's note: Dallas police actively guided an explosive-laden robot to kill gunman Micah Johnson when negotiations broke down after he opened fire on police and others July 7. This interview on Think -- which was recorded before the ambush took place -- focuses on the questions raised by computers and robots programmed to kill without human supervision. You may have seen it earlier this year when a company called Boston Dynamics posted a video of a humanoid robot that can walk on two legs, even over uneven terrain. Seeing a machine balance on bumpy, snowy ground in a New England forest is mesmerizing. But watching the robot get knocked to the ground by a human tester, then get up all by itself and keep going is somehow profoundly unsettling.

According to the American Cancer Society, cancer kills more than 8 million people each year. Early detection can boost survival rates. Researchers and clinicians are feverishly exploring avenues to provide early and accurate diagnoses, as well as more targeted treatments. Blood screenings are used to detect many types of cancers, including liver, ovarian, colon and lung cancers. Current blood screening methods typically rely on affixing biochemical labels to cells or biomolecules.

OK, let's drop autonomous vehicles for a second. This post is about creating a machine that writes its own code. More specifically, we are going to train a character level Long Short Term Memory neural network to write code itself by feeding it Python source code. The training will run on a GPU instance on EC2, using Theano and Lasagne. If some of the words here sound obscure to you, I will do my best to explain what is happening. This experiment is greatly inspired by this awesome blog post that I highly recommend reading.