Google I/O 2017 culminated Friday, but tech circuits are still abuzz about the many announcements that came out of the conference. Several new products, features within products and initiatives were announced at Google I/O and should be prominent throughout the rest of 2017. As with many Google conferences, the tech giant announced the latest figures of its continued growth, which include over 2 billion active users on Android and over 1 billion monthly users on each of its major platforms, including Google Search, Android, Play Store, Gmail, Chrome, YouTube and Google Maps. Here's a rundown of the other announcements and details shared during Google I/O 2017. Google Lens is essentially a search engine, which uses images and videos instead of words.

Ford CEO Mark Fields has "retired," pushed out by the board of directors after 28 years with the company. The surprise announcement of Jim Hackett as his replacement offers a clearest indication that while investors and car companies alike obsess over "mobility," they still have no idea what precisely that means. Fields, 56, devoted half his life to trying to make Ford successful. He started his career in the company's marketing arm in California, before moving to help manage their South American operations. He served a stint in Japan, too, helming Mazda while it was under Ford's control.

A computer wasn't supposed to be able to beat a grandmaster at the ancient game of Go for at least another decade. But AlphaGo, an artificially intelligent system designed by Google-owned DeepMind, did just that. In its public debut last year at a tournament in Seoul, AlphaGo thrashed Lee Sedol, the best player of last decade. Now AlphaGo is back, facing off in China against the world's top player to show just how much further machine-approximated intuition has advanced over the past year, and WIRED is there. Tomorrow morning, AlphaGo is set to play 19-year-old Ke Jie in Wuzhen, a town crisscrossed by canals 80 miles west of Shanghai.

Executives from three media companies shared data strategies and insights at the 2017 INMA World Congress on Monday. "We're very lucky at the South China Morning Post to be at the very start of this transformation," said Chief Executive Editor Gary Liu. The company recognises data as an asset to treasure, Lui said, but added that it is only an asset if it is taken advantage of. "Being able to capture and collect that data doesn't do us any good if we don't know how to utilise it," he said. Liu then gave a brief overview of the components driving SCMP's data strategy.

China's biggest online retailer, JD.com Inc., announced plans Monday to develop drone aircraft capable of carrying a ton or more for long-distance deliveries. The company said it will test the drones on a network it is developing to cover the northern Chinese province of Shaanxi. It said they will carry consumer goods to remote areas and farm produce to cities. JD.com, headquartered in Beijing, says it made its first deliveries to customers using smaller drones in November. Other e-commerce brands including Amazon.com Inc. also are experimenting with drones for delivery. "We envision a network that will be able to efficiently transport goods between cities, and even between provinces, in the future," the chief executive of JD's logistics business group, Wang Zhenhui, said in a statement.

This productivity improvement also decreased the price of the Ford Model T and increased production tenfold to around 300,000 cars. That's more than the company's 300 competitors built with four times the number of employees. Critically, however, Ford's increase in productivity also drove a rapid growth in the business that created more jobs and almost doubled wages. In 1914 Henry Ford raised the average pay in his factories from $2.54 to $5 per day. Thanks to the moving assembly line, more people worked and made more money.

Many big practices like Riverside Medical Group are looking forward to what most pharmaceutical companies call "beyond the pill." This is a phrase that has already become very popular among US and European pharmaceutical companies. This is because most drug makers are interested less on developing technologies with pills than with non-pill solutions. For example, lipo surgery has become a lasting solution to that of weight-loss pills. Although the insert molding process for the medical industry will continue to be of use in the production and packaging of existing pills, pharmaceutical companies are less likely to become aggressive in bringing in new pills.

The "world's first operational Robocop" started its tour of duty Sunday in Dubai – the first in the emirate's planned robot police force, according to the Daily Mirror. The Robocop, five feet five inches tall and weighing 220 pounds, speaks six languages and reads facial expressions. "He can chat and interact, respond to public queries, shake hands and offer a military salute," Brigadier-General Khalid Nasser Al Razzouqi, Director-General of Smart Services with the Dubai Police told the Mirror. Residents can use the Robocop to pay fines or report crimes, and it also can transmit and receive messages from police headquarters. The launch "is a significant milestone for the Emirate and a step towards realizing Dubai's vision to be a global leader in smart cities technology adoption," Al Razzouqi said.

Security, privacy, and fairness have become critical in the era of data science and machine learning. More and more we see that achieving universally secure, private, and fair systems is practically impossible. We have seen for example how generative adversarial networks can be used to learn about the expected private training data; how the exploitation of additional data can reveal private information in the original one; and how what looks like unrelated features can teach us about each other. Confronted with this challenge, in this paper we open a new line of research, where the security, privacy, and fairness is learned and used in a closed environment. The goal is to ensure that a given entity (e.g., the company or the government), trusted to infer certain information with our data, is blocked from inferring protected information from it. For example, a hospital might be allowed to produce diagnosis on the patient (the positive task), without being able to infer the gender of the subject (negative task). Similarly, a company can guarantee that internally it is not using the provided data for any undesired task, an important goal that is not contradicting the virtually impossible challenge of blocking everybody from the undesired task. We design a system that learns to succeed on the positive task while simultaneously fail at the negative one, and illustrate this with challenging cases where the positive task is actually harder than the negative one being blocked. Fairness, to the information in the negative task, is often automatically obtained as a result of this proposed approach. The particular framework and examples open the door to security, privacy, and fairness in very important closed scenarios, ranging from private data accumulation companies like social networks to law-enforcement and hospitals.

Ridesourcing platforms like Uber and Didi are getting more and more popular around the world. However, unauthorized ridesourcing activities taking advantages of the sharing economy can greatly impair the healthy development of this emerging industry. As the first step to regulate on-demand ride services and eliminate black market, we design a method to detect ridesourcing cars from a pool of cars based on their trajectories. Since licensed ridesourcing car traces are not openly available and may be completely missing in some cities due to legal issues, we turn to transferring knowledge from public transport open data, i.e, taxis and buses, to ridesourcing detection among ordinary vehicles. We propose a two-stage transfer learning framework. In Stage 1, we take taxi and bus data as input to learn a random forest (RF) classifier using trajectory features shared by taxis/buses and ridesourcing/other cars. Then, we use the RF to label all the candidate cars. In Stage 2, leveraging the subset of high confident labels from the previous stage as input, we further learn a convolutional neural network (CNN) classifier for ridesourcing detection, and iteratively refine RF and CNN, as well as the feature set, via a co-training process. Finally, we use the resulting ensemble of RF and CNN to identify the ridesourcing cars in the candidate pool. Experiments on real car, taxi and bus traces show that our transfer learning framework, with no need of a pre-labeled ridesourcing dataset, can achieve similar accuracy as the supervised learning methods.