It emerged earlier this month that thousands of Amazon employees are reviewing some Alexa recordings (which are captured after you've said the wake word). The auditors transcribe, annotate and analyze a selection of commands to help improve Alexa. But it seems these workers could view users' personal information too, according to Bloomberg. At least some employees are said to have had access to location data, addresses and phone numbers. There's no indication any workers have tried to look up a customer's home (say, on Google Street View) using the data from these tools.

With increasing physicians' workload and patients' needs for care, there is a need for technology that facilitates physicians work and performs continues follow-up with patients. Existing approaches focus merely on improving patient's condition, and none have considered managing physician's workload. This paper presents an initial evaluation of a conversational agent assisted coaching platform intended to manage physicians' fatigue and provide continuous follow-up to patients. We highlight the approach adapted to build the chatbot dialogue and the coaching platform. We will particularly discuss the activity recommender algorithms used to suggest insights about patients' condition and activities based on previously collected data. The paper makes three contributions: (1) present the conversational agent as an assistive virtual coach, (2) decrease physicians workload and continuous follow up with patients, all by handling some repetitive physician tasks and performing initial follow up with the patient, (3) present the activity recommender that tracks previous activities and patient information and provides useful insights about possible activity and patient match to the coach. Future work focuses on integrating the recommender model with the CoachAI platform and test the prototype with patient's in collaboration with an ambulatory clinic.

How to Do It is Slate's sex advice column. Send your questions for Stoya and Rich to howtodoit@slate.com. I have been sexually active since I was 17. I am now 29 years old. A majority of the sex I had between 17 and 21 was only when I was drunk, so I don't remember most of it, but I know I didn't climax.

There are three quite distinct ways to train a machine learning model on recommender system logs. The first method is to model the reward prediction for each possible recommendation to the user, at the scoring time the best recommendation is found by computing an argmax over the personalized recommendations. This method obeys principles such as the conditionality principle and the likelihood principle. A second method is useful when the model does not fit reality and underfits. In this case, we can use the fact that we know the distribution of historical recommendations (concentrated on previously identified good actions with some exploration) to adjust the errors in the fit to be evenly distributed over all actions. Finally, the inverse propensity score can be used to produce an estimate of the decision rules expected performance. The latter two methods violate the conditionality and likelihood principle but are shown to have good performance in certain settings. In this paper we review the literature around this fundamental, yet often overlooked choice and do some experiments using the RecoGym simulation environment.

You now have access to a treasure trove of government info through your smart speaker if you live in the UK. The British government has made over 12,000 pieces of Gov.uk information available through Alexa and Google Assistant, saving you the trouble of wading through official pages. Some of them are simple questions like the next bank holiday, while others are more involved questions such as obtaining a passport. Not everything is available, so you can't completely depend on a voice assistant just yet. However, there are promises of expansion.

Remember last week when Samsung unveiled its "foldable" phone? Well, it appears there are still a few wrinkles to iron out. Meanwhile, Google walkout organizers say they're facing retaliation from the company, a new Game of Thrones episode has come, and John Legend is putting Siri to shame. Here's the news you need to know in two minutes or less. Two Google employees who worked to organize a walkout of thousands of employees last November say the company is now retaliating against them.

Facebook has confirmed a report from earlier today saying it's working on an artificial intelligence-based digital voice assistant in the vein of Amazon's Alexa and Google Assistant. The news, first reported by CNBC, indicates Facebook isn't giving up on a vision it first put out years ago, when it began developing an AI assistant for its Messenger platform simply called M. This time around, however, Facebook says it is focusing less on messaging and more on platforms in which hands-free interaction, via voice control and potentially gesture control, is paramount. "We are working to develop voice and AI assistant technologies that may work across our family of AR/VR products including Portal, Oculus and future products," a Facebook spokesperson told The Verge today, following the initial report. That means Facebook may not position the product as a competitor to Alexa or similar platforms, but as more of a feature exclusive to its growing family of hardware devices. CNBC reported that the team building the assistant is working out of Redmond, Washington under the direction of Ira Snyder, a general manager at Facebook Reality Labs and a director of augmented and virtual reality at the company.

The inevitable march toward automation continues, analysts from the McKinsey Global Institute and from Tata Communications wrote in separate reports this week. Artificial intelligence's growth comes as no surprise -- a survey from Narrative Science and the National Business Research Institute conducted earlier this year found that 61 percent of businesses implemented AI in 2017, up from 38 percent in 2016 -- but this week's findings lay out in detail the likely socioeconomic impacts in the coming decade. The McKinsey models predict that 70 percent of companies will adopt at least one form of AI -- whether computer vision, natural language, virtual assistants, robotic process automation, or advanced machine learning -- by 2020. And Tata found unbridled enthusiasm among business leaders for an AI-dominated future; in a survey of 120 of them, 90 percent said they expect AI to enhance decision-making. McKinsey and Tata both contend that's a good thing.

The inevitable march toward automation continues, analysts from the McKinsey Global Institute and from Tata Communications wrote in separate reports this week. Artificial intelligence's growth comes as no surprise -- a survey from Narrative Science and the National Business Research Institute conducted earlier this year found that 61 percent of businesses implemented AI in 2017, up from 38 percent in 2016 -- but this week's findings lay out in detail the likely socioeconomic impacts in the coming decade. The McKinsey models predict that 70 percent of companies will adopt at least one form of AI -- whether computer vision, natural language, virtual assistants, robotic process automation, or advanced machine learning -- by 2020. And Tata found unbridled enthusiasm among business leaders for an AI-dominated future; in a survey of 120 of them, 90 percent said they expect AI to enhance decision-making. McKinsey and Tata both contend that's a good thing.

Beacon is unlike any other member of staff at Staffordshire University. It is available 24/7 to answer students' questions, and deals with a number of queries every day – mostly the same ones over and over again – but always stays incredibly patient. That patience is perhaps what gives it away: Beacon is an artificial intelligence (AI) education tool, and the first digital assistant of its kind to be operating at a UK university. Staffordshire developed Beacon with cloud service provider ANS and launched it in January this year. The chatbot, which can be downloaded in a mobile app, enhances the student experience by answering timetable questions and suggesting societies to join.