This week we begin with an article on explainable-AI, algorithms that can explain the thought processes behind their decision, and how they can help humans innovate better using a step-by-step feedback loop. Next is a report on the need for open policing data — and insights on the police involvement in marginalized communities — to improve accountability, build trust in the criminal justice system, and address racial inequities. The following article is on how Mastercard uses data exhaust to create new data-based products that help their customers, and ways other data companies can think of creating value of data. Following this is research on how the cancer database, a compilation of the most comprehensive characterization of head and neck cancers, could lead to new precision medicine treatments. Next, we have an analysis of the rise of AI in media and content production, the threat of deepfakes to liberal democracy, and technical solutions to tackle deepfakes and help build a resilient society. To end, we have an article on how a developer, using metadata, has compiled (in the form of an interactive map) archived videos uploaded by Parler users during the Capitol Hill siege.
How Explainable-AI Can Help Humans Innovate
THE FIELD OF ARTIFICIAL INTELLIGENCE (AI) has created computers that can drive cars, synthesize chemical compounds, fold proteins and detect high-energy particles at a superhuman level. However, these AI algorithms cannot explain the thought processes behind their decisions. A computer that masters protein folding and also tells researchers more about the rules of biology is much more useful than a computer that folds proteins without explanation.
Open Data: A Critical Tool for Police Reform and Racial Equity
Last week’s insurrection of far-right extremists storming the Capitol spurred lawmakers to consider important questions about the state of American democracy and the smooth transition of presidential power. The nature of the police response, and the contrast with police treatment of Black Lives Matter protesters, has also intensified concerns about racial bias in policing that have been building since the killing of George Floyd last May.
Data Exhaust Turbocharges Mastercar
Thirty years ago, consultants Stan Davis and Bill Davidson wrote a book called 2020 Vision in which, not surprisingly, they laid out their vision what would happen by 2020. I am not quite ambitious enough to review all their predictions, but one is of particular interest to me. In fact, in a Wall Street Journal column 7 years ago, I wrote this about one prescient Davis and Davidson prediction: a company’s “information exhaust” (information byproducts gathered in the course of its normal business) could be used to “informationalize” a business (develop products and services based on information).
Cancer Database Paves the Way for Precision Medicine Therapies
Researchers from the Johns Hopkins departments of oncology and pathology, the Johns Hopkins Kimmel Cancer Center, the Johns Hopkins University (JHU) School of Medicine, and 18 other organizations around the US and Poland have compiled a database of head and neck cancers to speed the development of precision medicine therapies.
Artificial Intelligence Breakthrough: Most Videos to be Made by AI in ‘Less Than a Decade’
Nina Schick, the author of ‘Deep Fakes and the Infocalypse’, made the claim. AI algorithms are producing increasingly advanced synthetic, or fake, videos some of which are being used to spread political disinformation. Speaking to Express.co.uk Ms Schick said: “We are increasingly facing a future where media and content production is going to be made by AI. So the future is completely synthetic. Some experts who I interviewed for my book said that within five to seven years’ time 90 percent of video content online is going to be synthetic.”
Developer Makes Interactive Map of Parler Videos from Capitol Hill Riots
A developer calling themselves Patr10tic has taken archived versions of videos uploaded by Parler users during the deadly Capitol Hill siege, geolocated them, reuploaded them, and placed them on an interactive map for anybody to watch.
Source: https://mailchi.mp/zigram/data-asset-weekly-dispatch_18_january_3
