Inside | Real news, curated by real humans
Inside AI

Inside AI (Aug 12th, 2019)

1. Lucidworks raised an additional $100 million in funding that will help it grow its AI engine for personalized search services. The company, which has raised around $200 million to date, received the latest investment from PE firm Francisco Partners and TPG Sixth Street Partners. CEO Will Hayes said Lucidworks' platform differs from those used by Amazon, Google, and Microsoft in how it uses AI to personalize search results after “sorting through mountains of data." Lucidworks builds algorithms that utilize other data sources, such as a user's location and previous shopping history. - TECHCRUNCH

2. The U.S. government should pursue "deeper, consistent, long-term engagement” of AI, according to a newly released plan from the National Institute of Standards and Technology (NIST). The White House issued an executive order back in February directing the NIST to create the plan, which outlines proposed standards and suggests tools to advance AI in the U.S. The plan nine identified areas of focus for AI standards: concepts and terminology, data and knowledge, human interactions, metrics, networking, performance testing and reporting methodology, safety, and risk management; and trustworthiness. - MERITALK

3. A recently unsealed deposition from 2016 claims that a victim of billionaire Jeffrey Epstein was forced to engage in sex with AI pioneer Marvin Minsky, a member of MIT’s Artificial Intelligence Lab. Minsky, who died in 2016, pioneered early self-training algorithms, outlining them in his 1969 book "Perceptrons," and founded MIT's AI Project, a precursor to its Computer Science and Artificial Intelligence Laboratory (CSAIL), in the 1950s. Victim Virginia Giuffre testified that in the early 2000s, she was directed to have sex with Minsky, then 73, when she was a 17-year-old at Epstein’s compound on his private island in the U.S. Virgin Islands. Maxwell’s attorney denied the allegations and called them “salacious and improper.” - THE VERGE

4. Political scientist Baobao Zhang penned an op-ed for The New York Times about the government's responsibility to expand a social safety net during the AI revolution. By 2075, Zhang says, there is a 90 percent chance that machines will perform better than people in most economically relevant tasks. Zhang writes that while most economists and technologists agree that AI and robotics are rapidly reshaping our economy, there is a need for a new American dream in a world "where hard work, grit, and ability are not enough to make a living." While machines have freed people from dangerous or boring jobs, they've also curbed the need for human labor at a faster pace "than new, decent-paying jobs are being created," writes Zhang, a Ph.D. candidate at Yale University. - NY TIMES

5. A chatbot under development at the University of Alberta will be geared toward helping people with neurodevelopmental disabilities. Leading the project is a team made up of AI specialists from the Alberta Machine Intelligence Institute, social scientists, and AI education experts from Canada. The goal is to create a chatbot that would direct users - in this case, people with NDD or their family members - to top sites on Google, as well as actually mimic conversations with health, education, and social science experts. It can take anywhere from six months to a year, after the initial referral, for children with a disability to see a specialist. - TECH XPLORE

6. Online AI moderators — including Google's Perspective — showed bias against posts written in African-American English or by people who identify as African American, according to research by NLP student Maarten Sap and his colleagues at the University of Washington. Perspective and other AIs trained to detect abusive or hateful content were nearly twice as likely to identify the African-American tweets as offensive, the researchers found. - NEW SCIENTIST

7. An AI-powered app is helping banana farmers scan for diseases or pests to prevent widespread outbreaks. The app called Tumaini — which means "hope" in Swahili — was trained on 20,000 images depicting banana disease and pest symptoms and uses image-recognition technology. Users scan the fruit and the app can label the disease and tell them the steps to take to address it, as well as uploading data, including location, into a larger global system for disease monitoring and control. - PHYS.ORG

8. A Machine-Read Analyst Sentiment, or “MRAS,” developed by Morgan Stanley's research team was able to beat the S&P 500 by an average of 1.9 percent over a 60-day period. The bank, which unveiled the strategy in June, trained the AI using over 40,000 reports to develop sentiment scores for analysts' notes. As CNBC describes, the team utilized the sentiment scores in the reports "with a price target change for a stock to take hypothetical positions." - CNBC

9. Huawei has launched its first HarmonyOS device, a smart TV called Honor Vision. Honor Vision is a “smart screen” product that looks similar to a 55-inch TV, but is equipped with Huawei's Wi-Fi chipset and a pop-up camera with facial recognition, body tracking and posture detection. It also shares images 100 times faster than Bluetooth and can control other smart home devices from the TV screen. - MASHABLE

This story first appeared in Inside IoT.

10. Tech entrepreneur Bruno Aziza recommends these three books for brushing up on your AI knowledge. They are "The Big Nine: How the Tech Titans and Their Thinking Machines Could Warp Humanity," by Future Today Institute founder Amy Webb; "Only Humans Need Apply: Winners and Losers in the Age of Smart Machines," by author Tom Davenport; and Stephen Hawking's perennial "Brief Answers to the Big Questions." - FORBES

Written and curated by Beth Duckett in Orange County. Beth is a former reporter for The Arizona Republic who has written for USA Today, Get Out magazine and other publications. Follow her tweets about breaking news and other topics in southern California here.

Editor: Kim Lyons (Pittsburgh-based journalist and managing editor at Inside).

Subscribe to Inside AI