This article was first published in the July 2015 issue of WIRED magazine. Be the first to read WIRED's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online The future of artificial intelligence begins with a game of Space Invaders. From the start, the enemy aliens are making kills -- three times they destroy the defending laser cannon within seconds. Half an hour in, and the hesitant player starts to feel the game's rhythm, learning when to fire back or hide. Finally, after playing ceaselessly for an entire night, the player is not wasting a single bullet, casually shooting the high-score floating mothership in between demolishing each alien. No one in the world can play a better game at this moment. This player, it should be mentioned, is not human, but an algorithm on a graphics processing unit programmed by a company called DeepMind. Instructed simply to maximise the score and fed only the data stream of 30,000 pixels per frame, the algorithm -- known as a deep Q-network – is then given a new challenge: an unfamiliar Pong-like game called Breakout, in which it needs to hit a ball through a rainbow-coloured brick wall. "After 30 minutes and 100 games, it's pretty terrible, but it's learning that it should move the bat towards the ball," explains DeepMind's cofounder and chief executive, a 38-year-old artificial-intelligence researcher named Demis Hassabis. "Here it is after an hour, quantitatively better but still not brilliant. But two hours in, it's more or less mastered the game, even when the ball's very fast. After four hours, it came up with an optimal strategy -- to dig a tunnel round the side of the wall, and send the ball round the back in a superhuman accurate way. The designers of the system didn't know that strategy."

Over the last several years, artificial intelligence (AI) has shifted from being an esoteric branch of computer science to an everyday technology that most of us carry in a pocket or purse--AI is what drives Apple's Siri, Facebook's photo-tagging, Spotify playlists and Google's auto-complete, just for starters. But can we also expect that someday soon AI will report and write the important news of the day--and technology stories like this one? Well, guess what: It already has. First, a bit of background: Many of the most exciting AI advances are driven by research in cognitive computing and natural language generation (NLG) processing, which allow computers to analyze massive quantities of data and generate a plain English document that highlights the most important insights. Those advances are made stronger through deep learning, a field of AI that uses neural networks to teach computers to sift through massive amounts of data to find their own patterns.

What is the relationship between artificial intelligence and cybersecurity? One certainly strengthens the other. Thanks to "machine learning", which is already well known among researchers, AI makes it possible to tackle security in a different way, better suited to the changing context of cybercrime, with greater anticipation and through behavioural analyses. Applications have already been released by laboratories, particularly via open-source libraries, however new skills are required. "By improving its knowledge and understanding of the phenomena by itself, machine learning enables an attack to be detected even if it is not familiar with the signature."

Delivery by drone may be legal within two years. Just don't expect many pizzas or packages to wing their way through your neighborhood by then. Despite huge interest in unmanned aerial vehicles (UAVs), and considerable hype around the idea of using them to deliver goods, experts say significant challenges still need to be solved for drone delivery to get off the ground. Google and Amazon are leading the development of delivery drones, while UPS, FedEx, and a host of startups are also researching the technology. Earlier this month, the U.S. Senate Transportation Committee drafted a bill that paves the way for regulation of delivery drones within two years.

A new low-power, "brain-inspired" supercomputing platform based on IBM chip technology will soon start exploring deep learning for the U.S. nuclear program. Lawrence Livermore National Laboratory announced on Tuesday that it has purchased the platform, based on the TrueNorth neurosynaptic chip IBM introduced in 2014. It will use the technology to evaluate machine-learning and deep-learning applications for the National Nuclear Security Administration. The computer will process data with the equivalent of 16 million neurons and 4 billion synapses and consume roughly as much energy as a tablet PC. Also included will be an accompanying ecosystem consisting of a simulator; a programming language; an integrated programming environment; a library of algorithms and applications; firmware; tools for composing neural networks for deep learning; a teaching curriculum; and cloud enablement.

The radio spectrum is a mess: it's congested, expensive and there's no room for expansion. But DARPA has a plan to change that, by building a system where radio waves can work together using artificial intelligence, rather than fighting for space. DARPA launched its latest Grand Challenge last week, and it plans to encourage researchers around the world to develop "smart systems that collaboratively, rather than competitively, adapt in real time to today's fast-changing, congested spectrum environment... to maximize the flow of radio frequency". That sounds exciting, because making radio frequency flow more easily means -- theoretically, at least -- faster data rates, fewer dropped signals, and cheaper connections. How does it plan to do it?

Donald Trump's speeches are nothing like that, according to Brad Hayes, a MIT scientist who programmed a Twitter bot to sound like him. Called DeepDrumpf, it uses an artificial intelligence algorithm based on Trump's language in hundreds of hours of debate transcripts. Hayes told Tech Insider he has learned a lot over the last few weeks about how Trump talks. Here is how he describes Trump's language, which differs dramatically from past presidential candidates. Trump often uses short, imperative sentences, Hayes says.

Latency to end-users and regulatory requirements push large companies to build data centers all around the world. The resulting data is "born" geographically distributed. On the other hand, many machine learning applications require a global view of such data in order to achieve the best results. These types of applications form a new class of learning problems, which we call Geo-Distributed Machine Learning (GDML). Such applications need to cope with: 1) scarce and expensive cross-data center bandwidth, and 2) growing privacy concerns that are pushing for stricter data sovereignty regulations. Current solutions to learning from geo-distributed data sources revolve around the idea of first centralizing the data in one data center, and then training locally. As machine learning algorithms are communication-intensive, the cost of centralizing the data is thought to be offset by the lower cost of intra-data center communication during training. In this work, we show that the current centralized practice can be far from optimal, and propose a system for doing geo-distributed training. Furthermore, we argue that the geo-distributed approach is structurally more amenable to dealing with regulatory constraints, as raw data never leaves the source data center. Our empirical evaluation on three real datasets confirms the general validity of our approach, and shows that GDML is not only possible but also advisable in many scenarios.

IBM is about to deliver the foundation of a brain-inspired supercomputer to Lawrence Livermore National Laboratory, one of the federal government's top research institutions. The delivery is one small "blade" within a server rack with 16 chips, dubbed TrueNorth, and is modeled after the way the human brain functions. Silicon Valley is awash in optimism about artificial intelligence, largely based on the progress that deep learning neural networks are making in solving big problems. Companies from Google to Nvidia are hoping they'll provide the AI smarts for self-driving cars and other tough problems. It is within this environment that IBM has been pursuing solutions in brain-inspired supercomputers. The main benefit is that such chips may be able to operate at lower frequencies and get much more work done on a much smaller amount of power.

Microsoft has created a new chat bot to "learn" from the internet… but she picked up a lot of bad habits. The tech company announced the launch of Tay this week, an artificial intelligence bot that is learning to talk like millennials by analysing conversations on Twitter, Facebook and the internet. The company's optimistic techies explained: "Tay is an artificial intelligent chat bot developed by Microsoft's Technology and Research and Bing teams to experiment with and conduct research on conversational understanding. "Tay is designed to engage and entertain people where they connect with each other online through casual and playful conversation. The more you chat with Tay the smarter she gets."