This week we saw some interesting M&A activity. Microsoft made yet another acquisition in the field of AI. This time they acquired Lobe, who’s developed an AI drag and drop tool. In addition we saw U.K. based Medopad acquire Sherbit to continue to enhance their AI driven disease detection solutions. KPMG has previously said Medopad is an AI unicorn in the making, so will be interesting to see how they progress post acquisition.
Company developments:
BMW drives to cut battery costs, share costs on autonomous vehicles: executive – Sept. 15, 2018 (Reuters)
- German luxury vehicle maker BMW AG (BMWG.DE) plans more deals with mining companies to secure electric vehicle battery materials, and is open to forming alliances to share the costs of developing autonomous-vehicle systems, the automaker’s research and development chief told Reuters
- BMW on Saturday showed off the electric, autonomous “Vision iNext” SUV in Beijing as part of a publicity tour in which the prototype car and a stage have traveled around the world in the hold of a Lufthansa (LHAG.DE) cargo jet. The vehicle, with gas and brake pedals that sink into the floor during autonomous driving and touch-screen controls embedded in the rear seat cushion, hints at an electric vehicle BMW could launch among 12 fully electric models it has promised by 2025
Oklahoma City stores will deliver groceries with autonomous vehicles – Sept. 14, 2018 (Engadget)
- Next year, Oklahoma City residents will be able to have their groceries delivered to them by an autonomous vehicle. Udelv announced this week that a new partnership will bring its self-driving delivery vehicles to the city’s largest local chain of grocery stores, which includes supermarkets such as Uptown Grocery, Buy For Less, Buy For Less Super Mercado and Smart Saver. Ten vehicles are scheduled to be delivered to the stores by the end of June 2019
- Udelv made its first delivery with the vehicles in California this January, and since then, it has completed more than 700 deliveries in partnership with a handful of merchants in the San Francisco Bay Area. The company and its Oklahoma City parter Esperanza Real Estate Investments will work with city authorities ahead of the vehicles’ deployment and Oklahoma’s Secretary of Transportation, Mike Patterson, said in a statement that the state has a regulatory group in place focusing on the use of autonomous delivery vehicle technology
AI Expert Leaves Google Over Plan To ‘Capitulate’ To China Censorship – Sept. 14, 2018 (Android Headlines)
- Machine learning scientist Jack Poulson quit his senior researcher position at Google over the company’s plans to “capitulate” to China, i.e. the Far Eastern country’s censorship and surveillance practices, the industry veteran told The Intercept. The Ph.D. holder resigned due to the existence of Project Dragonfly, a widely reported initiative seeking to return Google Search to China nearly a decade after the firm discontinued the Chinese version of its engine in order to protest Beijing’s censorship
Drone startup Airware crashes, shut downs after burning $118M – Sept. 14, 2018 (TechCrunch)
- Drone operating system startup Airware today suddenly informed employees it will cease operations immediately despite having raised $118 million from top investors like Andreessen Horowitz, Google’s GV, and Kleiner Perkins. The startup ran out of money after trying to manufacture its own hardware that couldn’t compete with drone giants like China’s DJI. The company at one point had as many as 140 employees, all of which are now out of a job
- Airware makes a cloud sofware system that helps enterprise customers like construction companies, mining operations, and insurance companies reviewing equipment for damages to use drones to collect and analyze aerial data. That allowed companies to avoid using expensive helicopters or dangerous rigs with humans on harnesses to make inspections and gauge work progress
Volvo’s self-driving cars will soon hit the streets in Sweden – Sept. 14, 2018 (The Local)
- The Swedish Transport Agency (Transportstyrelsen) recently granted the Swedish auto giant permission to begin real-world testing of its self-driving cars, boosting the company’s chances of meeting its stated goal of bringing the technology to market by 2021
- Transport Agency documents obtained by Di Digital revealed that the self-driving cars will be allowed on motorways and streets in the Gothenburg area. The cars will not be allowed to exceed 60 kilometres per hour when the self-driving features are activated and drivers are required to keep at least one hand on the wheel at all times. Operators of the autonomous vehicles must also undergo training provided by Volvo
Nasdaq Planning to Add Crypto Price Forecasting Tools, Uses NLP & Machine Learning – Sept. 13, 2018 (Coin Idol)
- Nasdaq, the globe’s second-biggest stock exchange, is purportedly considering adding tools to its Analytics Hub which will aid users in forecasting the price trends of Cryptos. Introduced in 2017, Nasdaq’s Analytics Hub applies Natural Language Processing (NLP) and machine learning to ascertain information on social media platforms
Nvidia unveils Tesla T4 chip for faster AI inference in datacenters – Sept. 12, 2018 (VentureBeat)
- Nvidia today debuted the Tesla T4 graphics processing unit (GPU) chip to speed up inference from deep learning systems in datacenters. The T4 GPU is packed with 2,560 CUDA cores and 320 Tensor cores with the power to process queries nearly 40 times faster than a CPU
- Analysis by Nvidia found that nearly half of all inference performed with the P4 in the span of the past two years was related to videos, followed by speech processing, search, and natural language and image processing
Apple’s autonomous vehicle fleet swells 27% in four months – Sept. 11, 2018 (TechCrunch)
- Over the past 18 months, Apple has gone from just three autonomous vehicles to 27 by January, 55 by May and now 70. GM Cruise has the most permitted autonomous test vehicles at 175, followed by Waymo with 88. Apple has the third-largest fleet
- The tech company’s permit with the CA DMV, the agency responsible for monitoring AVs in the state, is the only official acknowledgment that it even has a program. Apple’s self-driving program has been considered an open secret in Silicon Valley. CEO Tim Cook has more recently made references to the company’s interest in autonomous systems
Google’s AI Hate Speech Detection Tricked by Typos – Sept. 11, 2018 (PC Mag)
- As TNW reports, a group of researchers at Aalto University and the University of Padua discovered Google’s artificial intelligence can easily be tricked and that state-of-the-art hate speech detection models only perform well when tested by the same type of data on which they were trained. Simple tricks to get around Google’s AI include: inserting typos; adding spaces between words; or adding unrelated words to the original sentence
- Google’s method of hate speech detection is achieved through assigning a toxicity score to a piece of text, defining it as rude, disrespectful, or unreasonable enough that you would be inclined to leave the conversation. However, the AI system is not intelligent enough to detect the context of expletives, and a simple change between “I love you” and “I fucking love you” sees a change in score from 0.02 to 0.77
M&A:
Microsoft acquires Lobe, a drag-and-drop AI tool – Sept. 13, 2018 (TechCrunch)
- Microsoft today announced that is has acquired Lobe, a startup that lets you build machine learning models with the help of a simple drag-and-drop interface. Microsoft plans to use Lobe, which only launched into beta earlier this year, to build upon its own efforts to make building AI models easier, though, for the time being, Lobe will operate as before
- It’s worth noting that Lobe’s approach complements Microsoft’s existing Azure ML Studio platform, which also offers a drag-and-drop interface for building machine learning models, though with a more utilitarian design than the slick interface that the Lobe team built. Both Lobe and Azure ML Studio aim to make machine learning easy to use for anybody, without having to know the ins and outs of TensorFlow, Keras or PyTorch. Those approaches always come with some limitations, but just like low-code tools, they do serve a purpose and work well enough for many use cases
Medopad acquires U.S. rival Sherbit for AI-driven disease detection – Sept. 11, 2018 (VentureBeat)
- Medopad, a U.K. health technology startup that leverages machine learning to connect patients with a range of health care professionals, has announced its expansion into the U.S. with its acquisition of San Francisco-based Sherbit. Founded out of a Harvard Medical School in 2014, Sherbit is a similar proposition to Medopad in that it‘s focused on using personal data to uncover health insights through sensors, devices, and apps. This acquisition makes a great deal of sense for Medopad, as it seeks greater inroads into the U.S. market, but it’s also an unusual move in that it’s rare to see a U.K. company acquire a U.S. counterpart
- So far, Medopad has worked with a number of hospitals in the U.K., and it also has projects in Germany, Singapore, and China — indeed, Medopad was one of a number of companies that recently visited China with U.K. Prime Minister Theresa May to help foster relationships. Following that visit, Medopad secured $143 million worth of business in the country and raised $28 million in a series A round of funding
Fundraising / investment:
Integrate.ai pulls in $30M to help businesses make better customer-centric decisions – Sept. 12, 2018 (TechCrunch)
- Helping businesses bring more firepower to the fight against AI-fuelled disruptors is the name of the game for Integrate.ai, a Canadian startup that’s announcing a $30M Series A today. The round is led by Portag3 Ventures. Other VCs include Georgian Partners, Real Ventures, plus other (unnamed) individual investors also participating. The funding will be used for a big push in the U.S. market
- Its SaaS platform targets consumer-centric businesses — offering to plug paying customers into a range of AI technologies and techniques to optimize their decision-making so they can respond more savvily to their customers. Aka turning “high volume consumer funnels” into “flywheels”, if that’s a mental image that works for you
Chinese Video Content AI Start-Up Video++ Raises New Funding Round – Sept. 11, 2018 (China Money Network)
- Video++, a Shenzhen-based AI Start-up focused on maximizing the commercial potential of the online video industry, today announced that it had raised a C1 round of financing. New investors including Ubtech Robotics, Sirius Venture Partners, Wenxuan Equity Investment Fund, Real Power Capital, CV Capital and Hanfor Holdings participated in this round. No financial details were announced
- Founded in 2014, Video++ focuses on image recognition in video content, using computer vision technology to recognize places, objects, faces, brands and other elements in video content. These useful markers then help make video content searchable and placed in a structured format to better serve advertisers and e-commerce companies. Its revenue in 2017 reached RMB150 million (US$22 million) and is expected to exceed RMB500 million (US$73 million) in 2018. The revenue growth is mainly coming from advertising marketing and e-commerce businesses, the company says
Partnerships:
China’s DeepBlue Technology, Luxembourg LHoFT set up joint AI labs – Sept. 13, 2018 (Xinhua.Net)
- China’s leading artificial intelligence firm DeepBlue Technology and Luxembourg LHoFT, a fintech hub, signed an agreement Wednesday to open three joint labs in Luxembourg. The joint labs will focus on mutual openness and cooperation between China and Europe in the fields of basic research and application of artificial intelligence (AI)
- “The world is in the midst of great development, great changes and major adjustments in which countries are becoming more connected. Openness and cooperation are the only paths to help us cope with the global trends to build a new world pattern for AI,” Chen sai
Major UK news publisher to use AI in journalism after signing deal with London tech start-up – Sept. 11, 2018 (Compelo)
- Reach plc, previously known as Trinity Mirror, has struck a partnership with Krzana to use its artificial intelligence newsgathering platform in seven local newsrooms as a pilot. The bot can discover news – as defined by a journalist – from more than 60,000 sources worldwide, including social media, blogs, government website posts, businesses, organisations and individuals
- Reach, which publishes the Daily Mirror and The People alongside a portfolio of local and regional newspapers in Britain, will initially introduce the AI in journalism tool to newsrooms in cities such as Birmingham, Liverpool, Manchester, Newcastle and Cambridge. Users can set up highly detailed geographical parameters for searches to create a personal news feed, enabling identification of newsworthy events the second they happen
Research / studies:
Nvidia researchers generate synthetic brain MRI images for AI research – Sept. 16, 2018 (ZD Net)
- A group of researchers from Nvidia, the Mayo Clinic, and the MGH & BWH Center for Clinical Data Science this weekend are presenting a paper on their work using generative adversarial networks (GANs) to create synthetic brain MRI images. GANs are effectively two AI systems that are pitted against each other — one that creates synthetic results within a category, and one that identifies the fake results. Working against each other, they both improve
- In addition to widening the potential data sets, Shin and his colleagues say using GANs could provide a solution for the privacy challenges that surround the use of patient data. Because the synthetic images are not tied to a specific patient, it’s more anonymous and safer to transfer outside of a hospital
MIT CSAIL designs AI that can track objects over time – Sept. 14, 2018 (Venture Beat)
- Humans draw on an implicit understanding of the physical world to predict the motion of objects — and to infer interactions between them. If you’re presented with three frames showing toppling of cans — one with the cans stacked neatly on top of each other, the second with a finger at the stack’s base, and a third showing the cans lying on their sides — you might guess that the finger was responsible for their demise
- Robots struggle to make those logical leaps. But in a paper from the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory, researchers describe a system — dubbed a Temporal Relation Network (TRN) — that essentially learns how objects change over time. They aren’t the first to do so — Baidu and Google are among the firms who’ve investigated AI-assisted spatial-temporal modeling — but the team from MIT claim their method strikes a good balance between the accuracy and efficiency of previous approaches
Accelerating electrocatalyst discovery with machine learning – Sept. 13, 2018 (Phys.org)
- Researchers are paving the way to total reliance on renewable energy as they study both large- and small-scale ways to replace fossil fuels. One promising avenue is converting simple chemicals into valuable ones using renewable electricity, including processes such as carbon dioxide reduction or water splitting. But to scale these processes up for widespread use, we need to discover new electrocatalysts—substances that increase the rate of an electrochemical reaction that occurs on an electrode surface. To do so, researchers at Carnegie Mellon University are looking to new methods to accelerate the discovery process: machine learning
- While a human could study roughly 10 to 20 new energies a week, the machine can study hundreds per day. Prior to the automated system, researchers would have to narrow the space down to one class of materials and work in that space. Now, they can take a more holistic approach
A plan to advance AI by exploring the minds of children – Sept. 12, 2018 (MIT)
- Josh Tenenbaum, who leads the Computational Cognitive Science lab at MIT, is the head of a major new AI project called the MIT Quest for Intelligence. The project brings computer scientists and engineers together with neuroscientists and cognitive psychologists to explore research that might lead to fundamental progress in artificial intelligence
- Tenenbaum’s research focuses on exploring cognitive science in order to understand human intelligence. His work has, for example, explored how even small children are able to visualize aspects of the world using a kind of innate 3-D model. This gives humans greater instinctive understanding of the physical world than a computer or robot has. “Children’s play is really serious business,” he said. “They’re experiments. And that’s what makes humans the smartest learners in the known universe.”
MIT Lincoln Laboratory develops AI that shows its decision-making process – Sept. 11, 2018 (VentureBeat)
- In a paper published by MIT’s Lincoln Laboratory Intelligence and Decision Technologies Group (“Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning“) presented this summer at the Conference on Computer Vision and Pattern Recognition (CVPR), researchers describe a system that performs “human-like” reasoning to answer questions about the world — namely, the contents of images
- “Like workers down an assembly line, each module builds off what the module before it has figured out to eventually produce the final, correct answer,” according to MIT News. “As a whole, TbD-net utilizes one AI technique that interprets human language questions and breaks those sentences into subtasks, followed by multiple computer vision AI techniques that interpret the imagery.”
University of Minnesota to develop machine learning techniques for monitoring global change – Sept. 11, 2018 (Univ. of Minnesota)
- The University of Minnesota announced today that it has received a three-year, $1.43 million grant from the National Science Foundation to advance machine learning techniques to better monitor global agricultural and environmental change—a practice that can help society address the challenges of adapting to a changing climate, managing land use and natural resources, and sustainably feeding a growing population
- Machine learning, where computers “learn” from the data they collect without additional manual programming, is an effective tool for analyzing information about earth systems from multiple sources across time and space to study how natural processes and human activities affect the planet’s physical landscape and environment. The NSF grant funds a team of researchers at the University’s College of Science and Engineering (CSE); College of Food, Agricultural and Natural Resource Sciences (CFANS); and Minnesota Supercomputing Institute (MSI) to advance the state-of-the-art in machine learning for analyzing data from earth-observing satellites and generating actionable information on a global scale
Government / policy:
Gemalto to upgrade Irish welfare fraud detection facial recognition system – Sept. 14, 2018 (Biometric Update)
- Ireland’s Department of Employment Affairs and Social Protection has awarded a €383,000 (US$447,00) contract to Gemalto for an upgrade of its facial verification system. The department has used its existing facial biometrics system to identify 28 cases of identity fraud so far in 2018, and has saved the government more than €4 million ($4.67 million) since the system’s launch in 2012, TheJournal.ie reports
- The facial imaging software saved €334,000 ($390,000) in the first half of 2018, according to a department spokesperson. The department estimates savings of €894,000 ($1.04 million) in 2017 and €1.734 million ($2.02 million) in 2016
Blackberry CEO says driverless car industry requires more government regulation – Sept. 14, 2018 (Beta Kit)
- “A car could easily be infected with viruses (and) is literally a fully loaded weapon. If hackers can get hold of it, you can imagine what they could do,” Chen said. He also said that environmental conditions can prove just as dangerous
- “Regulation, and safety and security tech needs to be established well before I think anyone should allow the cars on the road,” he said. “The self-driving car still has a lot of human error and safety control.” BlackBerry is currently developing software for driverless vehicles with Baidu, the Chinese web search giant
IBM Created Facial Recognition Software With Racial Profiling Options for NYPD – Sept. 13, 2018 (Security Sales)
- IBM secretly used video surveillance images collected by the New York Police Department (NYPD) to develop facial recognition software that can identify individuals based upon certain features, such as hair and skin color, The Intercept reported
- IBM created the software that allowed the NYPD to search for potential criminals based upon tags including facial features, clothing color, facial hair, skin color, age, gender and more. Overall, it could identify more than 16,000 data points making it a very accurate way of recognizing faces, according to The Intercept. However, officers were told not to use the filters such as skin color or tone in their assessments, but only for evaluation purposes. The project was abandoned in 2016 when the NYPD decided not to roll it out to officers
Federal Standard on Autonomous Vehicle Technology Is Needed, Rep. Latta Says – Sept. 13, 2018 (Transport Topics)
- Rep. Bob Latta (R-Ohio), sponsor of House-passed legislation on autonomous vehicles, emphasized the urgency in having Congress agree on a national road map on policy for autonomous cars and trucks. Such a road map would help improve safety and enhance freight connectivity, he argued
- Rep. Bill Shuster (R-Pa.), chairman of the House Transportation and Infrastructure Committee, has examined autonomous technology throughout his tenure. During a recent visit to Carnegie Mellon University in his home state, the chairman said he received an update on self-driving cars from their researchers. Four years ago, Shuster hosted Carnegie Mellon University researchers for a demonstration of their driverless car near his office on Capitol Hill.
Events
Gridsum Wins First Place at the 2018 China AI and Law Challenge – Sept. 12, 2018 (PR Newswire)
- Gridsum Holding Inc. (“Gridsum” or the “Company”) (NASDAQ :GSUM ), a leading provider of cloud-based big-data analytics and artificial intelligence (“AI”) solutions in China, announced today that it has won first place in the 2018 Chinese AI and Law Challenge (“CAIL2018”). CAIL2018 was both initiated and led by the Information Centre of the Supreme People’s Court and the Central Youth Development Department of the Communist Youth League of China, and co-organized by various organizations, including the China Justice Big Data Institute, Chinese Information Processing Society of China, CETC Youth League committee, Tsinghua University, Peking University, and Institute of Software, Chinese Academy of Sciences
- CAIL2018 facilitates academic exchange and promotes the use of natural language processing and AI in the legal sector. There were more than 600 teams from both China and overseas participating in CAIL2018. Participants included engineers from top academic institutions including Tsinghua University and Peking University and internationally renowned companies such as Google and Microsoft. At CAIL2018, judges assessed contestants on their analysis of criminal case datasets and their application of AI across three categories including crime prediction, legislative recommendation, and judgement outcomes prediction
Intel announces winner of its AI Interplanetary Challenge – Sept. 11, 2018 (VentureBeat)
- Artificial intelligence won’t just teach robots to manipulate objects and improve the accuracy of pneumonia diagnoses — with a little ingenuity and elbow grease, it could help cosmologists unravel the mysteries of the universe. Intel today announced the winner of its AI Interplanetary Challenge, a two-phase, 12-week contest that tasked software developers, students, and data scientists with applying artificial intelligence to “space-related” problems
- The grand prize: a three-hour lunch with Planetary Society CEO Bill Nye (of Bill Nye Science Guy fame) and board member and actor Robert Picardo (Star Trek)
YiXue Education Squirrel AI founder joins world-leading experts at global summit to highlight AI’s impact on the future of Chinese education – Sept. 11, 2018 (PR Newswire)
- Founder of YiXue Squirrel AI Learning Derek Haoyang Li took part in a breakout session on the theme of Youth & the Future of Learning at Singularity University’s Global Summit held 20-22 Aug. in San Francisco. Singularity University’s 10th annual summit brought together 1600 education experts, enterprises, entrepreneurs and social and government leaders from 64 countries to discuss breakthroughs in innovation, EdTech and the future of education
- The high-profile founder of YiXue Squirrel AI Learning – he was named this year in both China Entrepreneur Magazine’s “Top 21 Stars of the Future” and Fast Company’s “China’s 100 Most Creative People” – was joined on stage at the summit by Esther Wojcicki of exploration-based learning company Planet3, Hla Hla Win of Myanmar-based education innovation firm 360ed and Arvind Nandakumar of spoken-word inspired youth group Youth Speaks. The panel was hosted by Gary A. Bolles, Chair for the Future of Work at Singularity University