Shift Technology, a software-as-a-service (SaaS) platform that leans on artificial intelligence (AI) to help companies combat insurance fraud, has raised 10 million in a round led by Accel, with participation from existing investors Iris Capital and Elaia Partners. Founded out of Paris in 2013, Shift Technology taps machine learning smarts to combat insurance fraud, using what it calls a "decision support platform" that automates the process of detecting fraud and helps humans prioritize which cases to follow-up on. In addition to using big data to highlight spurious claims, Shift Technology's tools also explain the kind of fraud that's likely involved and point to specific facets of a claim that are worth exploring further. As with most machine-learning systems, Shift Technology promises to improve over time by tapping user feedback and additional data. The company claims to have processed more than 50 million claims for insurers around the world since it was launched in 2014.

Havas extends its relationship with IBM, launching Havas Cognitive. Havas Group has been working with IBM for over 20 years now (early on Havas was one of the founding members of IBM's Marketing Innovation Group and an early Watson Ecosystem partner), but the relationship is continuing to evolve, with a new practice designed to help brands tap into artificial intelligence data and use it more effectively. TD Ameritrade and Adidas have been early pioneers of the technology, called Havas Cognitive. In the past, Havas has worked more closely with IBM as a client, but the new initiative represents more of a partnership between IBM and the Havas Group. The partnership came about when the two companies realized that they could create a real game-changing product for clients if they worked together.

With much of what Google announced during the first day of its I/O developer conference focused on helping users answer questions before they even think of them, artificial intelligence is proving critical for the company's strategy. "Google is betting the farm on artificial intelligence (A.I.) and they need to because Microsoft and Facebook aren't too far behind," said Patrick Moorhead, an analyst with Moor Insights & Strategy. "A.I. will determine the next-generation experience with all electronics. It essentially predicts what users want before they know they do. A.I. is the next big frontier, and Google has always been a pioneer in A.I." During the keynote speech on Wednesday, the company unveiled several upcoming products, including its Google Assistant, Google Home device, the Allo chat app and the Duo video chat app.

Rutgers is taking a leading role in an IBM-sponsored World Community Grid project that will use supercomputing power to identify potential drug candidates to cure the Zika virus. The project, known as OpenZika, employs a global team of scientists who will perform "virtual" experiments in a search of treatments for the fast-spreading virus that the World Health Organization has declared a global public health emergency. OpenZika will screen current drugs and millions of drug-like compounds from existing databases against models of Zika protein structures (and also against structures of proteins from related viruses, including West Nile Virus and Dengue). These computational results will be shared quickly with the research community and general public, with compounds showing the most promise then tested in laboratory settings. "Instead of having to wait a number of years, even decades potentially, to test all these compounds in order to find a few that could form the basis of antiviral drugs to cure Zika, we will perform these initial tests in a matter of months, just by using idle computing power that would otherwise go to waste," says Alex Perryman, a research teaching specialist at Rutgers' New Jersey Medical School, in Professor Joel Freundlich's lab.

TOKYO (Reuters) – Japan's Sony Corp said it plans to build up its artificial intelligence business and eventually turn it into a major revenue source, beginning with an investment in a U.S. startup. The electronics maker has invested an undisclosed sum in California-based Cogitai. The year-old firm, founded by three researchers, focuses on technology that allows machines to learn continually and autonomously from interaction in the real world. The move comes a time when major technology companies such as Facebook Inc, Apple Inc and Alphabet Inc's Google are spending aggressively on AI ventures. "From an objective perspective, we are lagging behind," Chief Executive Hiroaki Kitano at Sony Computer Science Laboratories said in an interview. "But there are still unexplored areas – some in cyberspace but vastly more in the physical world," Kitano said.

Google I/O is the biggest date in the firm's calendar: it's the conference where it gets everyone together in one big room (actually, this year it was a tent) in California to reveal all the cool stuff it has been working on. This year was more focused than most. It's the first I/O since the company restructured itself into Alphabet, hiving off the wilder projects such as Google Glass, self-driving cars and life-extension research into other subsidiaries. Gone are the days of livestreaming a parachute jumper wearing Google Glass from the I/O stage. But that doesn't mean the company wasn't trying to impress with more down-to-earth revelations.

As part of an onslaught of upcoming products, Google will implant a more personable form of artificial intelligence into an Internet-connected device called Home, which echoes the Echo, Amazon.com's Meanwhile, Google will also delve deeper into the still-nascent realm of virtual reality with a system called Daydream that's meant to challenge Facebook-owned Oculus' early lead in fabricating artificial worlds. In an attempt to outshine Apple, Google is also adding features to its Android operating system, including the ability to run apps without actually installing them on a device. That feature, called Instant Apps, might have been the biggest breakthrough that Google announced Wednesday at its annual developers conference held in an amphitheater a few blocks from its Mountain View, California, headquarters. It is the first time that Google has held the conference in its hometown since the inaugural event in 2006.

Google prides itself on the safety record of its groundbreaking self-driving cars but as an extra precaution the company is working on a "sticky" hood that will prevent pedestrians from bouncing off the cars if hit. Having driven over 1.5 million miles since beginning testing in 2014, Google's self-driving cars reported their first accident only last month when the search giant admitted it was at least partly responsible for an accident, which saw one of its fleet of Lexus SUVs collide with a public bus. Previously, all crashes reported involving one of Google's fleet had been blamed on driver error. Google is keen to continue leading the way in the area of autonomous cars amid competition from traditional carmakers as well as companies like Faraday Future, Tesla and reportedly Apple. To that end, Google wants to help protect those pedestrians who may wander in front of one of its cars -- by making them "stick" to the hood.

Everyone has a digital assistant these days. Microsoft has baked Cortana into the heart of all its software and Google has just announced a major revamp to what it now calls Google assistant. Apple, on the other hand, has seen its digital assistant Siri fall behind the competition in recent years, but the company is now set to make the biggest overhaul to the feature yet by bringing it to the desktop. The long-rumored move was given more credibility this week when MacRumors published leaked screenshots from a reliable source, showing Siri working on Mac OS X, the Apple desktop operating system that is due to be updated next month. The leaked images show a Siri icon in the menu bar as well as a recognizable waveform logo situated in the part of the interface known as the Dock.

Recently, there's been a great deal of excitement and interest in deep neural networks because they've achieved breakthrough results in areas such as computer vision.1 However, there remain a number of concerns about them. One is that it can be quite challenging to understand what a neural network is really doing. If one trains it well, it achieves high quality results, but it is challenging to understand how it is doing so. If the network fails, it is hard to understand what went wrong. While it is challenging to understand the behavior of deep neural networks in general, it turns out to be much easier to explore low-dimensional deep neural networks – networks that only have a few neurons in each layer.