Verizon said Wednesday it had turned on its ultrafast 5G wireless network in parts of Chicago and Minneapolis, though it will be available only to certain subscribers who pay a fee and own a compatible smartphone. The move makes Verizon the first wireless carrier in the United States to flip the switch on speedy, smartphone-ready 5G service in select urban areas, the company said in a statement, though other U.S. carriers including AT&T, Sprint and T-Mobile have pledged to do the same in the coming months. The service Verizon is offering -- 5G, the fifth generation of wireless data networks -- could provide consumers Internet speeds that are up to 100 times faster than 4G networks, according to an industry trade association. Through the placement of small boxes that serve as conduits for invisible, data-transmitting radio waves, 5G networks could power a wide range of consumer devices, from smartphones that can stream Netflix videos more quickly to enabling the arrival of self-driving cars. The promise of faster speeds and more reliable connections has generated a full-on race between AT&T, Verizon, Sprint and T-Mobile, the country's four largest carriers, to see who can offer service first (and capture new consumers and their cash in the process). The U.S. government also has taken notice, with lawmakers and the Trump administration looking to supercharge research, investment and development in the telecom sector, believing that better wireless networks will grant the country a more competitive business edge -- particularly against China, which is racing to deploy 5G as well.

In late 2018, digital readiness ranked as the top concern in enterprise C suites in a survey conducted by North Carolina State University's Poole College of Management. That's not a surprise considering that enabling digital readiness is more than just digitizing information. Work processes must be retooled to work in a digital environment, employees must be trained--and digitized information must be trusted. SEE: Digital transformation in 2019: A business leader's guide to future challenges and opportunities (Tech Pro Research) What best approaches and projects should IT leaders pursue to facilitate digital transformation? CRM, sales, products and service systems should be cross-linked so that personnel in customer-facing departments have complete and consistent information on customers.

The US Food and Drug Administration has proposed a framework on how it might regulate medical devices that rely on AI and machine learning algorithms. The report published this week outlines two types of algorithms for the purposes of regulation: "locked algorithms" and "adaptive algorithms." Locked algorithms provide the same result each time they're fed the same input. The answers are normally based on things like look-up tables, decision trees, or classifiers. An adaptive algorithm, however, will "change its behavior using a defined learning process."

Think you've been scammed and are looking for help on USA.gov? "Sam" is here for you. Sam is a new automated chatbot built by a small team at USA.gov that can answer basic consumer questions like "Is this a scam?" "Can I get my money back?" and more. Sam's reach is limited thus far -- intentionally so -- in that it's only available from three webpages on USA.gov. But for the team behind it, Sam is an exciting initial pilot with what could prove to be a powerful technological tool.

According to Gartner Research, from now until the year 2022, 85 percent of AI projects will deliver erroneous outcomes due to bias in data, algorithms, or the teams responsible for managing them. This bias is something I've known for a while as I followed AI systems from the late 1980s. The fact of the matter is that people program and teach AI systems, so those AI systems will tend to have the innate biases of the people who teach them. The use of the cloud to host pricey AI systems is actually making things worse, because the number of companies that can afford AI has gone up but the number of people with solid AI skills has not grown at the same pace. So, in addition to that innate bias being in AI tools used more broadly, the lack of talent also means more mistakes in how the knowledge-bases are built are going to be common for some time.

Natural language processing (NLP) technologies leads the AI revolution. It is by far the most visible and popular form of AI. That's partly because it has to do with one of the most basic fundamental forms of intelligence -- language. With NLP, machines have made huge leaps in recent years in recognizing and understanding language -- and thus are now learning to communicate just like humans, whether it's a bot chatting with you via text, or Alexa speaking with you through an Amazon Echo. Up to a quarter of all companies will have integrated a virtual customer assistant or a chatbot -- using some form of NLP -- into their customer service by 2020, according to research firm Gartner.

When artificial intelligence systems start getting creative, they can create great things – and scary ones. Take, for instance, an AI program that let web users compose music along with a virtual Johann Sebastian Bach by entering notes into a program that generates Bach-like harmonies to match them. Run by Google, the app drew great praise for being groundbreaking and fun to play with. It also attracted criticism, and raised concerns about AI's dangers. My study of how emerging technologies affect people's lives has taught me that the problems go beyond the admittedly large concern about whether algorithms can really create music or art in general.

IBM receives more than 8,000 resumes a day, making it No. 1 on job-search site Glassdoor for Gen Z applicants, said IBM CEO Ginni Rometty at CNBC's @ Work Talent HR Summit on Tuesday in New York City. But that's not the only way the technology giant, which employs roughly 350,000 workers, knows who in the workforce is currently searching for a new position. IBM artificial intelligence technology is now 95 percent accurate in predicting workers who are planning to leave their jobs, said Rometty. During Rometty's seven-year tenure as CEO, IBM has been improving its AI work devoted to the retention of its employees. "The best time to get to an employee is before they go," she said.

Yoshua Bengio, who last month won the prestigious Turing award, alongside Geoffrey Hinton and Yann LeCun, for his work on AI, is worried about what the technology is being made into behind closed doors. In an interview with Nature, he explains his concerns, but takes care to avoid sounding like a doomsayer. A professor at the Montreal Institute for Learning Algorithms, his main concern is not a particular nightmare scenario but simply that AI is being pursued by people who have few controls in place. "A lot of what is most concerning is not happening in broad daylight," he said. This we have certainly seen, with all the major tech companies in one way or another providing or considering government and military work, from the benign to the clearly conflict-oriented.

In many real-world planning problems with factored, mixed discrete and continuous state and action spaces such as Reservoir Control, Heating Ventilation and Air Conditioning (HVAC), and Navigation domains, it is difficult to obtain a model of the complex nonlinear dynamics that govern state evolution. However, the ubiquity of modern sensors allows us to collect large quantities of data from each of these complex systems and build accurate, nonlinear deep neural network models of their state transitions. But there remains one major problem for the task of control - how can we plan with deep network learned transition models without resorting to Monte Carlo Tree Search and other black-box transition model techniques that ignore model structure and do not easily extend to mixed discrete and continuous domains? In this paper, we introduce two types of nonlinear planning methods that can leverage deep neural network learned transition models: Hybrid Deep MILP Planner (HD-MILP-Plan) and Tensorflow Planner (TF-Plan). In HD-MILP-Plan, we make the critical observation that the Rectified Linear Unit (ReLU) transfer function for deep networks not only allows faster convergence of model learning, but also permits a direct compilation of the deep network transition model to a Mixed-Integer Linear Program (MILP) encoding. Further, we identify deep network specific optimizations for HD-MILP-Plan that improve performance over a base encoding and show that we can plan optimally with respect to the learned deep networks. In TF-Plan, we take advantage of the efficiency of auto-differentiation tools and GPU-based computation where we encode a subclass of purely continuous planning problems as Recurrent Neural Networks and directly optimize the actions through backpropagation. We compare both planners and show that TF-Plan is able to approximate the optimal plans found by HD-MILP-Plan in less computation time. Hence this article offers two novel planners for learned deep neural net transition models: one optimal method for mixed discrete and continuous state and actions (HD-MILP-Plan) and a scalable alternative for large-scale purely continuous state and action problems (TF-Plan).