DETROIT - Ford revealed details of its long-awaited restructuring plan Monday as it prepared for a future of electric and autonomous vehicles by parting ways with 7,000 white-collar workers worldwide, about 10 percent of its global salaried workforce. The major revamp, which had been underway since last year, will save about $600 million per year by eliminating bureaucracy and increasing the number of workers reporting to each manager. In the U.S. about 2,300 jobs will be cut through buyouts and layoffs. About 1,500 have left voluntarily or with buyouts, while another 300 have already been laid off. About 500 workers will be let go starting this week, largely in and around the company's headquarters in Dearborn, Michigan, just outside Detroit.

NPR's Audie Cornish talks with Microsoft President Brad Smith about why he thinks the government should regulate artificial intelligence, especially facial recognition technology.

Recently, Convolutional Neural Networks (CNNs) demonstrate a considerable vulnerability to adversarial attacks, which can be easily misled by adversarial perturbations. With more aggressive methods proposed, adversarial attacks can be also applied to the physical world, causing practical issues to various CNN powered applications. Most existing defense works for physical adversarial attacks only focus on eliminating explicit perturbation patterns from inputs, ignoring interpretation and solution to CNN's intrinsic vulnerability. Therefore, most of them depend on considerable data processing costs and lack the expected versatility to different attacks. In this paper, we propose DoPa - a fast and comprehensive CNN defense methodology against physical adversarial attacks. By interpreting the CNN's vulnerability, we find that non-semantic adversarial perturbations can activate CNN with significantly abnormal activations and even overwhelm other semantic input patterns' activations. We improve the CNN recognition process by adding a self-verification stage to analyze the semantics of distinguished activation patterns with only one CNN inference involved. Based on the detection result, we further propose a data recovery methodology to defend the physical adversarial attacks. We apply such detection and defense methodology into both image and audio CNN recognition process. Experiments show that our methodology can achieve an average rate of 90% success for attack detection and 81% accuracy recovery for image physical adversarial attacks. Also, the proposed defense method can achieve a 92% detection successful rate and 77.5% accuracy recovery for audio recognition applications. Moreover, the proposed defense methods are at most 2.3x faster compared to the state-of-the-art defense methods, making them feasible to resource-constrained platforms, such as mobile devices.

The increasing accessibility of data provides substantial opportunities for understanding user behaviors. Unearthing anomalies in user behaviors is of particular importance as it helps signal harmful incidents such as network intrusions, terrorist activities, and financial frauds. Many visual analytics methods have been proposed to help understand user behavior-related data in various application domains. In this work, we survey the state of art in visual analytics of anomalous user behaviors and classify them into four categories including social interaction, travel, network communication, and transaction. We further examine the research works in each category in terms of data types, anomaly detection techniques, and visualization techniques, and interaction methods. Finally, we discuss the findings and potential research directions.

A recent hardware test of NASA's robotic assistant, 'Astrobees,' takes a new wave of space-bound autonomous helpers one step closer to reality. According to NASA, this month astronaut Anne McClain ran a hardware test of the robot, named'Bumble,' one of three robotic assistants launched to the International Space Station (ISS) on April 15. Scientists hope Bumble will carry out an array of housekeeping tasks like monitoring equipment and keeping inventory of supplies that NASA hopes will free up its astronauts to perform other more critical tasks relating to with their missions and experiments. Astrobees are just one of many robotic applications from NASA who is also studying the use of'soft' robotics that replace traditional hardware with malleable plastics'Astrobee will prove out robotic capabilities that will enable and enhance human exploration,' said Maria Bualat, Astrobee project manager at NASA's Ames Research Center in a statement. 'Performing such experiments in zero gravity will ultimately help develop new hardware and software for future space missions.'

Deep Learning is really starting to establish itself as a major new tool in visual effects. Currently the tools are still in their infancy but they are changing the way visual effects can be approached. Instead of a pipeline consisting of modelling, texturing, lighting and rendering, these new approaches are hallucinating or plausibly creating imagery that is based on training data sets. Machine Learning, the superset of Deep Learning and similar approaches have had great success in image classification, image recognition and image synthesis. At fxguide we covered Synthesia in the UK, a company born out of research first published as Face2Face.

Israeli radiology startup Aidoc has received FDA clearance for its AI-based product meant to help identify potential cases of pulmonary embolism in chest CT scans. Pulmonary embolism (PE) – which occurs when a blood clot gets lodged in the lung – is considered a silent killer that causes up to 200,000 deaths a year in the United States. The condition often strikes with little to no warning and diagnosis of a case can be extremely time-sensitive. Aidoc's technology doesn't require dedicated hardware and runs continuously on hospital systems, automatically ingesting radiological images. The 70-person company focuses on workflow optimization in radiology to help triage high risk patients for additional and faster review.

Over the last few years, India has taken significant steps towards adoption of emerging technologies like artificial intelligence (AI) and machine learning(ML), with technology solutions providers, tech leaders, startups and government agencies playing a significant role in shaping the evolution of the technology in the country. According to a recent study, which observed the country-wide AI readiness in the Asia Pacific, India was ranked third in readiness, with its overall readiness score being 50.2 out of100, while Singapore was ranked the first with 63 points and Hong Kong was at the second with 56.5 points. As straightforward as it might sound, AI readiness simply does not refer a country's preparedness in embracing AI, rather, a number of key factors like the ability of its consumers, businesses and government to adopt, deploy and support AI technologies are taken into consideration to better understand the readiness capability of a country in adopting AI. In other words, AI readiness is not a linear process instead, multiple factors shape the outcome. "AI adoption is fragmented and uneven across the region. In some cases, governments' efforts and commitment have yet to be reflected in businesses' or consumers' adoption and usage of AI. In others, business and consumers are taking the lead, showing governments the way forward in terms of change and innovation," Eric Loeb EVP, Global Government Affairs points out in the study.

If you board a flight out of the United States four years from now, chances are the government is going to scan your face -- an ambitious timeline that has privacy experts reeling. That's according to a recent Department of Homeland Security report, which says that U.S. Customs and Border Protection (CBP) plans to dramatically expand its Biometric Exit program to cover 97 percent of outbound air passengers within four years. Through this program, which was already in place in 15 U.S. airports at the end of 2018, passengers have their faces scanned by cameras before boarding flights out of the nation. If the AI-powered system determines that the photo doesn't match one on file, CBP officials can look into it. The goal of these airport face scans is purportedly to catch people who have overstayed their visas, but civil liberties expert Edward Hasbrouck sees them as potentially giving the government increased control over American citizens.

The Pentagon is one of the largest technology customers in the world, purchasing everything from F-35 planes (roughly $90 million each) to cloud services (the JEDI contract was $10 billion). Despite outlaying hundreds of billions of dollars for acquisitions though, the Defense Department has struggled to push nascent technologies from startups through its punishing procurement process. The department launched the Defense Innovation Unit a few years back as a way to connect startups into the defense world. Now, the military has decided to work even earlier to ensure that the next generation of startups can equip the military with the latest technology. Cambridge, Mass.-based MIT and the U.S. Air Force announced today they are teaming up to launch a new accelerator focused on artificial intelligence applications, with the Air Force committed to investing $15 million into roughly 10 MIT research projects per year.