America's manufacturing heyday is gone, and so are millions of jobs, lost to modernization. Despite what Treasury Secretary Steven Mnuchin might think, the National Bureau of Economic Research and Silicon Valley executives, among many others, know it's already happening. And a new report from PwC estimates that 38% of American jobs are at "high risk" of being replaced by technology within the next 15 years. But how soon automation will replace workers isn't the real problem. The real threat to American jobs will come if China does it first.

OTTAWA – The Trudeau government's drive to transform Canada into an artificial-intelligence superpower is stirring warnings about the possible dark sides of a technology with vast – and largely unknown – potential. Among the concerns that surfaced this week: "killer robots." Known more formally as lethal autonomous weapons systems, the spectre of deadly-efficient technology came up during an open caucus meeting of Senate Liberals who were exploring the pros and cons of AI and robotics. Expert panellists at the meeting each nodded to the positives associated with these technologies, but there were also calls for Ottawa to develop a well-defined national AI plan. "The same underlying AI that can be engines of creation can also be engines of destruction," Ian Kerr, who holds a Canada Research Chair in ethics, law and technology at the University of Ottawa, told senators Wednesday in his opening statement.

Uber is putting the brakes on its driverless car pilot program after one of its self-driving cars got into a high speed crash in Arizona. Sean Dowling (@seandowlingtv) has more. A group of self driving Uber vehicles position themselves to take journalists on rides during a media preview at Uber's Advanced Technologies Center in Pittsburgh. SAN FRANCISCO -- A potentially pivotal lawsuit between Uber and Alphabet's Waymo over allegedly stolen self-driving vehicle sensor technology just took another screeching turn. Anthony Levandowski, a former Google employee who founded the now Uber-owned self-driving truck company Otto, invoked his Fifth Amendment right to protect himself from self-incrimination Wednesday, according to a transcript of the private court hearing reviewed Thursday by USA TODAY.

SAN FRANCISCO, Calif., March 30, 2017 – Cloudera, the provider of the leading global platform for machine learning and advanced analytics built on the latest open source technologies, announced accelerated momentum in the cybersecurity market. Over the past six months, Cloudera has earned multiple industry awards, advanced its cybersecurity solution based on Apache Spot (incubating), hosted a nine-city cybersecurity roadshow, and expanded its growing customer and partner base. Cloudera's cybersecurity solution provides enterprise-grade machine learning and advanced analytics to protect the hyper-connected enterprise. Security operations centers can no longer bolt on additional applications without a future-proof architecture that meets today's and tomorrow's needs. Cloudera validated this at this year's RSA conference when it surveyed 125 attendees and discovered the following: As evidence of the company's growing cybersecurity market leadership, Cloudera has received prominent industry nods for leadership and innovation, advanced open source technologies, and continuing customer and partner adoption: "We're honored by the awards we've received and thrilled by the traction we've made in the cybersecurity industry. Both are testament to the work we've done to help combat global adversaries," said Tom Reilly, chief executive officer at Cloudera.

Google is contributing to the $150 million going into The Vector Institute, a new initiative housed at the University of Toronto, guided by chief scientific adviser Geoffrey Hinton. The Institute is getting started today, with significant investment from the governments of Canada and the province of Ontario. Google's contribution is around $5 million, but it's likely the future of the two organizations will be closely aligned, since Hinton also acts as VP and Engineering Fellow at Google, where he has had a key role in spearheading AI via Google Brain. The Vector Institute is a dedicated AI research facility, and will use its amassed resources to fund research by postgraduate researchers working on projects in the field. The areas of focus for The Vector Institute include healthcare, financial business, plus manufacturing and material science. Meanwhile, Google is also launching Google Brain Toronto, which is actually the second Google Brain satellite office based in Canada.

America's biggest tech companies are remaking the internet through artificial intelligence. And more than ever, these companies are looking north to Canada for the ideas that will advance AI itself. This morning, Google announced it's starting an AI lab in Toronto. At the same time, it's helping to fund a public-private partnership with the University of Toronto to develop and commercialize AI talent and ideas. In November, the company made a similar move in Montreal--a city that has also attracted Microsoft's attention.

We must recognize that our cyberdefense technologies are not working and will not work. Cases in point: Our most sensitive cyberoffense technologies have been hacked; power companies admit they would have great difficulty stopping a cyberattack and are being asked to be prepared to operate at much less than full capacity under a cyberattack; 70 percent of oil and gas companies have been attacked -- and the threat is growing. The cybersecurity industry is in chaos and needs to move toward new technologies -- cyberdefense technologies that are beginning to leverage analytics, machine learning and artificial intelligence (AI). Hackers are taking advantage of the same technologies, so the cyberdefense industry needs to jump on board. Let's quit playing catch-up and instead take a proactive approach to cybersecurity.

Many applied settings in empirical economics involve simultaneous estimation of a large number of parameters. In particular, applied economists are often interested in estimating the effects of many-valued treatments (like teacher effects or location effects), treatment effects for many groups, and prediction models with many regressors. In these settings, machine learning methods that combine regularized estimation and data-driven choices of regularization parameters are useful to avoid over-fitting. In this article, we analyze the performance of a class of machine learning estimators that includes ridge, lasso and pretest in contexts that require simultaneous estimation of many parameters. Our analysis aims to provide guidance to applied researchers on (i) the choice between regularized estimators in practice and (ii) data-driven selection of regularization parameters. To address (i), we characterize the risk (mean squared error) of regularized estimators and derive their relative performance as a function of simple features of the data generating process. To address (ii), we show that data-driven choices of regularization parameters, based on Stein's unbiased risk estimate or on cross-validation, yield estimators with risk uniformly close to the risk attained under the optimal (unfeasible) choice of regularization parameters. We use data from recent examples in the empirical economics literature to illustrate the practical applicability of our results.

Foodborne illness afflicts 48 million people annually in the U.S. alone. Over 128,000 are hospitalized and 3,000 die from the infection. While preventable with proper food safety practices, the traditional restaurant inspection process has limited impact given the predictability and low frequency of inspections, and the dynamic nature of the kitchen environment. Despite this reality, the inspection process has remained largely unchanged for decades. CDC has even identified food safety as one of seven ”winnable battles”; however, progress to date has been limited. In this work, we demonstrate significant improvements in food safety by marrying AI and the standard inspection process. We apply machine learning to Twitter data, develop a system that automatically detects venues likely to pose a public health hazard, and demonstrate its efficacy in the Las Vegas metropolitan area in a double-blind experiment conducted over three months in collaboration with Nevada’s health department. By contrast, previous research in this domain has been limited to indirect correlative validation using only aggregate statistics. We show that adaptive inspection process is 64 percent more effective at identifying problematic venues than the current state of the art. If fully deployed, our approach could prevent over 9,000 cases of foodborne illness and 557 hospitalizations annually in Las Vegas alone. Additionally, adaptive inspections result in unexpected benefits, including the identification of venues lacking permits, contagious kitchen staff, and fewer customer complaints filed with the Las Vegas health department.

This issue features expanded versions of articles selected from the 2016 AAAI Conference on Innovative Applications of Artificial Intelligence held in Phoenix, Arizona. We present a selection of three articles that describe deployed applications, two articles that discuss work on emerging applications, and an article based on the 2016 Robert S. Engelmore Memorial Lecture.