Finland is one of the first countries to try out the minibuses on city roads thanks to its laws allowing cars to roam without a driver. Dubai had signed a deal with the company back in April to test the EasyMile vehicles, while a Japanese mall began using them to shuttle shoppers around this month. But neither of those will likely rival the live-traffic demands of the Helsini experiment. The buses won't be doing extensive hauls: the EZ10 model is built for short-range travel, say for ferrying folks between a metro station and bus stop, at a max speed of a little over six miles per hour. If all goes well, the vehicles will supplement but not replace existing mass transit networks.

Apple, Google, Facebook, and Microsoft are worldwide technology powerhouses, but when it comes to the adoption of virtual digital assistants (VDAs) like Siri, Google Assistant, and Cortana, scale only takes you so far. In this particular business, players who successfully cater to the nuances of individual countries will conquer the global VDA market. The same principle will apply to enterprises looking to automate customer interactions like customer service and e-commerce with enterprise VDAs. The challenges facing VDA providers were brought to light recently by the plight of Jibo, the crowdfunded smart home VDA robot that received pre-orders from consumers in 47 countries. On August 9, the company announced that product rollouts would be limited to the United States and Canada only, and that all orders for Jibo outside those markets will be refunded.

The era of big data has witnessed a paradigm shift into analytics. Today, it's no longer sufficient to simply gather data from social media, IoT, and wearable devices, and be unable to manage or filter it. It is more about delivering the right data to the right person, at the right time. This trend is growing crucial as data is multiplying every day and pouring in from various devices and smart machines including wearables, electronic gadgets, and other devices. Such factors call for the treatment of vast pools of structured and unstructured data with care and precision. This is precisely where invisible analytics come in.

Drone footage is offering a first look at a site where treasure hunters are digging for a buried Nazi train that could contain 250million worth of gold and other riches. With the dig gaining headlines around the world, the site in Walbrzych, Poland, has been sealed off to the prying eyes of the public, and the only way to get a glimpse of it is from high above. There is no guarantee the so-called'Nazi gold train' is buried there or even exists, but two explorers believe it is nearly 30ft below ground in a railway tunnel. Workers have set up fences and privacy screens to keep the public's prying eyes away In addition to gold and gems, Piotr Koper and Andreas Richter believe the site may have been used to hide the bodies of thousands of forced labourers. It could take up to 10 days for the site to be excavated.

A key component driving the development of driverless cars is machine learning and other artificial intelligence capabilities along with computer vision approaches used for image and signal processing. Ford Motor Co., which is targeting fully autonomous vehicles for ride sharing by 2021, unveiled a series of machine learning and machine vision deals as it doubles the size of it Silicon Valley research campus. The U.S. carmaker (NYSE: F) announced an acquisition and others investments on Tuesday (Aug. Ford also disclosed a licensing deal with machine vision specialist Nirenberg Neuroscience, who is credited with cracking the code the eye uses to transmit visual information to the brain. Furthering its autonomous vehicle initiative, Ford also announced an investment in the 3-D mapping startup, Civil Maps.

This AI Can Tell if You're Depressed by Looking at Your Instagram Photos AI, Bots and Canvases Part III: Gates and Ballmer paved the path for Nadella's AI and bots NYC Startup Init.ai is Putting the Intelligence Back into AI

The second-oldest confirmed shipwreck in the Great Lakes, an American-built, Canadian-owned sloop that sank in Lake Ontario more than 200 years ago, has been found, a team of underwater explorers said Wednesday. The three-member western New York-based team said it discovered the shipwreck this summer in deep water off Oswego, in central New York. Images captured by a remotely operated vehicle confirmed it is the Washington, which sank during a storm in 1803, team member Jim Kennard said. "This one is very special. We don't get too many like this," said Mr. Kennard, who along with Roger Pawlowski and Roland "Chip" Stevens has found numerous wrecks in Lake Ontario and other waterways.

This June, we published a big story--Shane Bauer's account of his four-month stint as a guard in a private prison. That's "big," as in XXL: 35,000 words long, or 5 to 10 times the length of a typical feature, plus charts, graphs, and companion pieces, not to mention six videos and a radio documentary. It was also big in impact. More than a million people read it, defying everything we're told about the attention span of online audiences; tens of thousands shared it on social media. The Washington Post, CNN, and NPR's Weekend Edition picked it up.

The MegaFace dataset contains 1 million images representing more than 690,000 unique people. It is the first benchmark that tests facial recognition algorithms at a million scale.University of Washington In the last few years, several groups have announced that their facial recognition systems have achieved near-perfect accuracy rates, performing better than humans at picking the same face out of the crowd. But those tests were performed on a dataset with only 13,000 images -- fewer people than attend an average professional U.S. soccer game. What happens to their performance as those crowds grow to the size of a major U.S. city? University of Washington researchers answered that question with the MegaFace Challenge, the world's first competition aimed at evaluating and improving the performance of face recognition algorithms at the million person scale.

Ford is finally taking the plunge and setting a date for the launch of a fleet of fully autonomous cars. Ford CEO Mark Fields announced that the company is working toward launching a fleet of commercial, level 4 (one level below a completely autonomous system, in which drivers don't have to be engaged) vehicles in a ride-hail service by 2021. The details of that ride-hail service -- such as which company Ford will partner with to operate it -- still haven't been determined. Fields made the announcement at a meeting with Ford employees at the company's Palo Alto research and development center. As part of that effort, Ford is investing in Velodyne, a self-driving tech company, and is working with three other startups.