Facial recognition technology's failures when it comes to accurately identifying people of color have been well documented and much criticized. But an attempt by Google to improve its facial recognition algorithms by collecting data from people with dark skin is raising further concerns about the ethics of the data harvesting. Google has been using subcontracted workers to collect face scans from members of the public in exchange for $5 gift cards, according to a report from the New York Daily News. The face scan collection project had been previously reported, but anonymous sources described unethical and deceptive practices to the Daily News. The subcontracted workers were employed by staffing firm Randstad but directed by Google managers, according to the report.

In this study, the authors discuss and show how new kinds of digital data and analytics methods and tools falling under the umbrella term of Big Data, including Artificial Intelligence (AI) systems, can help measure development effectiveness. Selected case studies provide examples of assessments of the effectiveness of ODA-funded policies and programmes. They use different data and techniques. For example, analysis of mobile phone data and satellite images: to estimate poverty and inequality, traffic congestion, social cohesion or machine learning approaches to social media analysis to understand social interactions and networks, and natural language processing to study changes in public awareness. A toolkit contains resources and suggestions on key steps and considerations, including legal and ethical, when designing and implementing projects aimed at measuring development effectiveness through new digital data and tools.

This year, self-driving cars started getting pretty good. Deep fakes video started getting pretty convincing. Our virtual assistants got to the point where they could understand us well enough to do some simple things, like tell us the weather or get driving directions home. When it comes to artificial intelligence, we have reached an inflection point. The technology is good enough to use. Next year promises to be a breakout year for AI, as it starts to permeate all aspects of our lives. Here are predictions for 2020 from some of the world's top AI experts. Jen Snell is VP of product marketing at Verint, where she leads a product strategy team focused on intelligent self-service, conversational AI, automation, and analytics. She is a frequent speaker and a leading contributor on topics shaping the development and design of interactive technologies. Follow her on Twitter @JenniferLSnell and on LinkedIn.

The AI for Services network brings together Data and Artificial Intelligence businesses and academics with professionals working in the high value service sectors of legal, accountancy insurance and finance. This initiative is part of the Industrial Strategy Next Generation Services Challenge fund programme. Contact Astrid Ayel if you would like more information on the network or to discuss collaboration opportunities; you can also get in touch with her via LinkedIn or Twitter. About us: What is the AI for Services Network? Find out more information about the Industrial Strategy Next Generation Services Challenge fund here.

There is currently keen interest amongst tech companies, investors, and research institutes in both artificial intelligence and blockchain technologies. The buzz has reached corporations and financial institutions looking to tailor these technologies for their own businesses in order to keep pace in our increasingly digitized economy. This excitement has implications for securing patent protection – Canada and the United States (as well as many other countries) operate on a "first to file" basis. That means you will want to file a patent application before other inventors monopolize protection for the technology in your area, potentially curtailing your competitive business advantage. The United States Patent and Trademark Office (USPTO) has issued revised guidelines on what constitutes subject-matter that is eligible for patent protection (under 35 U.S.C. § 101), effective January 7, 2019. These guidelines will likely have the largest impact in the area of computer-implemented inventions, including artificial intelligence and blockchain technologies, and may make it easier to patent some of these inventions.

On 6-7 May 2019, the World Economic Forum Centre for the Fourth Industrial Revolution and its partners UNICEF and the Canadian Institute for Advanced Research (CIFAR) hosted a workshop in San Francisco on the joint "Generation AI" initiative. This workshop identified deliverables in two key areas: 1) public policy guidelines that direct countries on creating new laws focused on children and 2) a corporate governance charter that guides companies leveraging AI to design their products and services with children in mind.

Merriam Webster: Definition of artificial intelligence a branch of computer science dealing with the simulation of intelligent behavior in computers the capability of a machine to imitate intelligent human behavior Techopedia explains Artificial Intelligence (AI) Artificial intelligence is a branch of computer science that aims to create intelligent machines. It has become an essential part of the technology industry. Research associated with artificial intelligence is highly technical and specialized. Artificial intelligence or AI refers to software technologies that make a robot or computer act and think like a human. Some software engineers say that it is only artificial intelligence if it performs as well or better than a human.

We are proud to share some words of the letter received from Borhene Chakroun, Director of Education Sector Division for policies and lifelong learning Systems (UNESCO). "In light of the very good results achieved by the above-mentioned Chair, confirmed by the...

This paper provides statistical theory and intuition for Personalized PageRank (PPR), a popular technique that samples a small community from a massive network. We study a setting where the entire network is expensive to thoroughly obtain or maintain, but we can start from a seed node of interest and "crawl" the network to find other nodes through their connections. By crawling the graph in a designed way, the PPR vector can be approximated without querying the entire massive graph, making it an alternative to snowball sampling. Using the Degree-Corrected Stochastic Blockmodel, we study whether the PPR vector can select nodes that belong to the same block as the seed node. We provide a simple and interpretable form for the PPR vector, highlighting its biases towards high degree nodes outside of the target block. We examine a simple adjustment based on node degrees and establish consistency results for PPR clustering that allows for directed graphs. We illustrate the method with the Twitter friendship graph and find that (i) the adjusted and unadjusted PPR techniques are complementary approaches, where the adjustment makes the results particularly localized around the seed node and (ii) the bias adjustment greatly benefits from degree regularization.

The concept of explainability is envisioned to satisfy society's demands for transparency on machine learning decisions. The concept is simple: like humans, algorithms should explain the rationale behind their decisions so that their fairness can be assessed. While this approach is promising in a local context (e.g. to explain a model during debugging at training time), we argue that this reasoning cannot simply be transposed in a remote context, where a trained model by a service provider is only accessible through its API. This is problematic as it constitutes precisely the target use-case requiring transparency from a societal perspective. Through an analogy with a club bouncer (which may provide untruthful explanations upon customer reject), we show that providing explanations cannot prevent a remote service from lying about the true reasons leading to its decisions. More precisely, we prove the impossibility of remote explainability for single explanations, by constructing an attack on explanations that hides discriminatory features to the querying user. We provide an example implementation of this attack. We then show that the probability that an observer spots the attack, using several explanations for attempting to find incoherences, is low in practical settings. This undermines the very concept of remote ex-plainability in general. 1 Introduction Modern decision-making driven by black-box systems now impacts a significant share of our lives [9, 29]. Those systems build on user data, and range from rec-ommenders [21] ( e.g., for personalized ranking of information on websites) to predictive algorithms ( e.g., credit default) [29]. This widespread deployment, along with the opaque decision process provided by those systems raises concerns about transparency for the general public or for policy makers [12]. This translated in some jurisdictions ( e.g., United States of America and Europe) into a so called right to explanation [12, 26], that states that the output decisions of an algorithm must be motivated. Explainability of in-house models An already large body of work is interested in the explainability of implicit machine learning models (such as neural network models) [2, 13, 20]. Indeed, those models show state-of-art performances when it comes to a task accuracy, but they are not designed to provide explanations -or at least intelligible decision processes-when one wants to obtain more than the output decision of the model. In the context of recommendation, the expression "post hoc explanation" has been coined [32].