Dr. T's AI brief
Daniel Tauritz
Trump Administration Requests $1 Billion For Non-Defense AI Research In 2020
U.S. News & World Report (9/26) reports that “AI has its own category in the president’s budget request for 2020, with about $1 billion sought in funding for non-defense purposes.” However, experts says this level of funding “is still not enough to maintain the United States’ competitive edge over China, which has been rapidly narrowing the gap in global AI research.” Still, the piece observes that the “budget request, published earlier this month by the National Science and Technology Council (NSTC), constitutes the ‘first-ever reporting of agency-by-agency federal investments’ in artificial intelligence research and development related to nondefense purposes.”
Former Google Engineer Warns AI May Accidentally Start A War
Newsweek (9/16) reports a former Google software engineer has warned that advancements in artificial intelligence “may result in ‘atrocities’ because the technology will behave in unexpected ways.” Computer scientist Laura Nolan “left Google in June last year after raising concerns about its work with the US Department of Defense on Project Maven, a drone program that was using AI algorithms to speed up analysis of vast amounts of captured surveillance footage.” Speaking to The Guardian, the software engineer “said the use of autonomous or AI-enhanced weapons systems that lack a human touch may have severe, even fatal, consequences.” She said, “What you are looking at are possible atrocities and unlawful killings even under laws of warfare, especially if hundreds or thousands of these machines are deployed. There could be large-scale accidents because these things will start to behave in unexpected ways.”
Army AI Task Force Builds New Prototypes For Armored Vehicles
Fox News (9/16, Osborn) reports streamlining multiple targeting sensors “to destroy long-range targets, arming forward-positioned robots to penetrate enemy defenses and receiving organized weather-specific terrain mapping from nearby drones – are all emerging combat dynamics increasingly made possible by AI-enabled weapons and technologies.” New applications of AI “are consolidating data from otherwise disparate sensor systems, analyzing seemingly limitless amounts of targeting data in seconds and instantly sifting through hours of drone video to massively improve attack options and shorten ‘sensor-to-shooter’ time.” Col. Doug Matty, Army AI Task Force Deputy Director, told Warrior in an interview, “We are developing an AI stack regarding how we pull together the sensors, computing layer and analytics to manage the data.” Matty explained, “We are focusing on the Next Gen Combat Vehicle. As you know they are progressing with two different platforms – the optionally-manned vehicle and the robotic combat vehicle.”
Washington Times Profiles US Army’s AI Task Force
The Washington Times (9/12, Wolfgang) reports the Army’s AI Task Force, created this year by the Defense Department, “has brought together military officials and academic leaders” at Carnegie Mellon University, “an institution considered to be the birthplace of AI research.” The Times says the AI Task Force is “a key piece of the Pentagon’s plan to beat China and Russia in the high-stakes artificial intelligence race.” The group “has been tasked with creating the weapons, vehicles and sensors needed for 21st-century combat” and is “feverishly working to incorporate AI into its operations.”
Amazon Among Companies Developing Deadly AI, Report Says
AFP (8/21) covers a new report from Dutch NGO Pax “that surveyed major players from the sector about their stance on lethal autonomous weapons.” Based on the report, AFP says “Amazon, Microsoft and Intel are among leading tech companies putting the world at risk through killer robot development.” The NGO “ranked 50 companies by three criteria: whether they were developing technology that could be relevant to deadly AI, whether they were working on related military projects, and if they had committed to abstaining from contributing in the future.” Lead author Frank Slijper asked, “Why are companies like Microsoft and Amazon not denying that they’re currently developing these highly controversial weapons, which could decide to kill people without direct human involvement?” AFP reports 21 “companies were of ‘medium concern,’ while 21 fell into a ‘high concern’ category, notably Amazon and Microsoft who are both bidding for a $10 billion Pentagon contract to provide the cloud infrastructure for the US military.”
Study Finds That Amazon, Microsoft Are ‘Putting World At Risk Of Killer AI’
International Business Times (8/21) says a recent report by Pax found that “Amazon, Microsoft, and Intel are among leading tech companies putting the world at risk through killer robot development.” The Dutch NGO surveyed “major players from the sector about their stance on lethal autonomous weapons,” ranking 50 companies by “three criteria: whether they were developing technology that could be relevant to deadly AI, whether they were working on related military projects, and if they had committed to abstaining from contributing in the future.” According to the report, 22 companies were of “‘medium concern,’ while 21 fell into a ‘high concern’ category, notably Amazon and Microsoft.”
Visa To Launch AI Testing Platform To Develop New Anti-Fraud Solutions
The Wall Street Journal (8/7, Castellanos, Subscription Publication) reports Visa is debuting a platform to help its engineers test AI algorithms for preventing credit-card fraud. The platform, set to launch later this year, serves as an example of the broader financial-services trend of utilizing AI technology to detect criminal behavior in transaction patterns. The banking sector is projected to be the second-largest spender on AI in 2019, behind only retail at $5.3 billion.
FDA Clears AI System For Detecting Collapsed Lung
AI in Healthcare (9/12, Pearson) reports the FDA has given 510(k) clearance to GE Healthcare’s Critical Care Suite, “a product that embeds AI algorithms in mobile X-ray machines.” It sends an alert to a physician “if it detects signs of a pneumothorax, a type of collapsed lung, as soon as a chest X-ray image is acquired.” GE Healthcare “says the development marks the first time the FDA has OK’d an AI application embedded in a medical device for flagging critical findings of this type.”
AI Streamlines Waste Sorting Process In China
China Daily (8/17, Shujuan) reports artificial intelligence is helping to “streamline the process” of waste sorting in China. On July 1st, the country announced “a plan to expand AI’s application in trash sorting as part of the city’s attempt to become a national and global leader in the technology.” With this development, China went onto announce “the application scenario of AI-powered autonomous waste sorting on July 2 along with the expansion of 27 other new AI applications.”
IBM’s AI-Powered E-Tongue Could Fight Food Fraud
Food Navigator (8/8, Askew) reports IBM Research is currently developing Hypertaste, “an electronic, AI-assisted tongue that, researchers say, draws inspiration from the way humans taste.” According to the company, Hypertaste “caters to a ‘growing need’ to identify liquids ‘swiftly and reliably’ without access to high-end laboratories.” The robot utilizes “combinational sensing” in order to efficiently “identify each separate component” of liquids. Of the development, IBM Research Lead Investigator of Hypertaste Patrick Ruch “told this publication that the technology could unlock the door to faster and more cost-effective testing to detect contaminants in the environment or food fraud.”
Yes Bank Partners With Microsoft To Strengthen AI-Enabled Chatbot
Livemint (IND) (9/4, Mathur) reports India’s Yes Bank is partnering with Microsoft to “strengthen its first of its kind, AI enabled chatbot, Yes Robot, with advanced neuro language programming (NLP) engine LUIS (Language Understanding Intelligent Service) and other cognitive services, capable of understanding and resolving the evolving banking needs of customers without the need for human intervention.” The updated Yes Robot is built on Microsoft’s “modular and customizable AI platform” and will allow users to “perform financial and non-financial banking transactions by employing conversational AI equipped with extensive financial knowledge, thereby relieving customers from the hassle of navigating through multiple web pages.”
Pentagon Releases AI Strategy
AP (2/12, O'Brien) reports the Pentagon on Tuesday “outlined its first AI strategy,” which “calls for accelerating the use of AI systems throughout the military, from intelligence-gathering operations to predicting maintenance problems in planes or ships.” The report “urges the U.S. to advance such technology swiftly before other countries chip away at its technological advantage.” However, it “makes little mention of autonomous weapons but cites an existing 2012 military directive that requires humans to be in control.”
Recogni Raises $25M For Self-Driving Car AI Hardware
SiliconANGLE (7/31) reports that Recogni Inc, a “startup developing artificial intelligence hardware for self-driving cars,” received backing from numerous large industry players in a $25 million round. Investment was led by GreatPoint Ventures, and included participation from Toyota Motor Corp, BMW AG, and Faurecia SA. Recogni “has built a hardware module that enables autonomous vehicles to see their surroundings” and is “based on a homegrown application-specific integrated circuit that turns raw data from a car’s sensors into information the navigation system can understand.”
VentureBeat (7/31) reports that Recogni CEO RK Anand said of the company’s technology, “The issues within the…autonomy ecosystem range from capturing [and] generating training data to inferring in real time. These vehicles need datacenter class performance while consuming minuscule amounts of power.” Anand added, “Leveraging our background in machine learning, computer vision, silicon, and system design, we are engineering a fundamentally new system that benefits the auto industry with very high efficiency at the lowest power consumption.”
Chinese Researchers Develop Autonomous Bike With Neuromorphic Chip
The New York Times (7/31, Metz) reports that a “team of researchers in China is rethinking autonomous transportation using a souped-up bicycle” which can roll over bumps on its own, turn on command, and swerve to encounter obstacles. The researchers “who built the bike believe it demonstrates the future of computer hardware” as it navigates “with help from what is called a neuromorphic chip, modeled after the human brain.” While “building the right hardware may require at least several more years of research,” the researchers “believe that time will bring far more than just autonomous bicycles” and paint the chip as “a step toward ‘artificial general intelligence.’”
Opinion: Bicycle That Learns Has A Lot To Teach Us About AI’s Future
Caixin Global (CHN) (8/2, Walsh, Murphy) reports on a team of Chinese and international scientists that have recently demonstrated the potential of “neuromorphic” hybrid chips by utilizing “one they developed to run an autonomous bicycle.” According to their findings in Nature, “the voice-activated, self-balancing two-wheeler can detect and avoid obstacles.” The chip – Tianjic – can “flip between a number of existing architectures to accommodate both the kind of machine-learning algorithms known to computer scientists, and the circuits modeled on animal brains familiar to neuroscientists.” The article notes that the chip’s creators are “not the first to experiment with neuromorphic technology” as the research arms of IBM and Intel are “also developing neuromorphic chips and devices.”
Tsinghua Researchers Develop Hybrid Chip For AI-Controlled Autonomous Bicycle
MIT Technology Review (8/1, Knight) reports that Luping Shi and colleagues at China’s Tsinghua University have developed a new chip called Tianjic that “features a hybrid design that seeks to bring together two different architectural approaches to computing: a conventional, von Neumann design and a neurologically inspired one.” In a paper (7/31, Pei) published in Nature, the researchers outline how the chip is used in an autonomous bicycle “to run artificial neural networks for obstacle detection, motor and balance control, and voice recognition, as well as conventional software.”
Medtronic Partners With Viz.ai To Bring AI Tech Into Stroke Care
MedCity News (8/1, Parmar) reports Medtronic is “aiming to leverage AI in another business: stroke care,” as the company announced last week “that it has entered into a global distribution agreement with Viz.ai, whose artificial intelligence-powered imaging software is aimed at quickly treating patients suspected of having ischemic issues.” Connected to CT scanners, Viz.ai can “quickly determine based on a CT scan of the patient’s brain, whether the patient has suffered a large-vessel occlusion, flag where it believes that occlusion has occurred and notify doctors.” MedCity calls the Viz.ai-Medtronic partnership “further proof that the largest pure-play medical device company believes that more in the stroke market are likely to buy into Viz.ai’s product and vision.”
Contributor: AI In Healthcare Needs To Focus On Causality
Forbes (8/14) contributor Alexander Lavin writes that there is “much to be excited about with artificial intelligence (AI) in healthcare,” yet “medical practitioners and researchers at the intersection of machine learning (ML) and medicine are quick to point out these successes are not representative of the more nuanced, non-trivial challenges presented by medical research and clinical applications.” Lavin adds, “Advances in causal machine learning will provide trials that are conducted completely within the confines of a computer.”
Analysis: Different AI Services Could Potentially Make Biggest Impact On US Medical Industries TIME (8/13, Hsu) reports on the optimism and caution over potential AI systems spreading throughout the medical industry. Currently, there are “different visions for how AI services could make the biggest impact,” and it’s “still unclear whether AI will improve the lives of patients or just the bottom line for Silicon Valley companies, health care organizations, and insurers.” TIME says if AI “works as promised, it could democratize health care by boosting access for underserved communities and lowering costs” as well as freeing “overworked doctors and reduce the risk of medical errors that may kill tens of thousands, if not hundreds of thousands, of U.S. patients each year.” Harvard Medical School biomedical informatics researcher Isaac Kohane says, “I think that all our patients should actually want AI technologies to be brought to bear on weaknesses in the health care system, but we need to do it in a non-Silicon Valley hype way.”