Ask poverty attorney Joanna Green Brown for an example of a client who fell through the cracks and lost social services benefits they may have been eligible for because of a program driven by artificial intelligence (AI), and you will get an earful. There was the "highly educated and capable" client who had had heart failure and was on a heart and lung transplant wait list. The questions he was presented in a Social Security benefits application "didn't encapsulate his issue" and his child subsequently did not receive benefits. "It's almost impossible for an AI system to anticipate issues related to the nuance of timing," Green Brown says. Then there's the client who had to apply for a Medicaid recertification, but misread a question and received a denial a month later.

Today's artificial intelligence systems, including the artificial neural networks broadly inspired by the neurons and connections of the nervous system, perform wonderfully at tasks with known constraints. They also tend to require a lot of computational power and vast quantities of training data. That all serves to make them great at playing chess or Go, at detecting if there's a car in an image, at differentiating between depictions of cats and dogs. "But they are rather pathetic at composing music or writing short stories," said Konrad Kording, a computational neuroscientist at the University of Pennsylvania. "They have great trouble reasoning meaningfully in the world." Original story reprinted with permission from Quanta Magazine, an editorially independent publication of the Simons Foundation whose mission is to enhance public understanding of science by covering research developments and trends in mathematics and the physical and life sciences.

AI robots and self-driving cars might steal the headlines, but the next big leap in technology will be advances in voice services, according to Google's head of search, Ben Gomes, who says that a better understanding of common language is crucial to the future of the internet. "Speech recognition and the understanding of language is core to the future of search and information," said Gomes . "But there are lots of hard problems such as understanding how a reference works, understanding what'he', 'she' or'it' refers to in a sentence. It's not at all a trivial problem to solve in language and that's just one of the millions of problems to solve in language." Gomes was speaking to the Guardian ahead of Google's 20th anniversary on 24 September, more than seven years after Google launched its first voice service as simple speech-to-text for search. Now built into Google's search and its AI voice assistant which is embedded in billions of smartphones around the globe, voice recognition has become essential in developing countries with low literacy rates.

Google looks to make a big splash at the 2018 Consumer Electronics Show, touting the Google Assistant. Apple TV has Siri, Amazon's Fire TV has Alexa, and now, Roku has joined forces with the Google Assistant to bring an established voice to its popular streaming players and branded TVs. Roku, the No. 1 streaming player, had offered its own voice search, but Google's Assistant, generally accessed via Google Home speakers, is more widely used by the public. Roku, in announcing new products for the fall Monday, didn't specify a time frame for the change, only saying it would be "soon," and for most existing devices. Additionally, the Roku TVs will have more functionality with Google, allowing viewers to say "Hey, Google," to turn their TV on and off, turn up the volume, mute, switch inputs and change channels, but only if the set is connected to an antenna.

Artificial intelligence promises an even bigger revolution than the internet yet could be stifled in the UK by a fear-driven public backlash, according to a leading scientist and broadcaster. Prof Jim Al-Khalili, a physicist and the incoming president of the British Science Association, warns that without greater transparency and public engagement the full potential of AI may not be realised. In the absence of concerted action by academics, the government and industry, the rapidly advancing technology could end up "uncontrolled and unregulated" in the hands of a few supremely powerful companies, he says. Previewing his presidential address at this year's British Science festival in Hull, which begins next week, Khalili spoke of the dream and dangers of AI. He said the UK was at the forefront of the technology, which is predicted to contribute up to $15tn (£11.7tn) to the global economy by 2030.

Of the many acronyms engineers spend their lives internalizing, few are more valuable than KISS: Keep It Simple, Stupid. Constrain the problem, reduce the variables, and make life as easy as possible when designing novel systems--like, say, a self-driving car. The world is a messy, complicated place. The less of it you need to solve, the closer you are to having a working product. That's why Waymo tests and plans to deploy its vehicles in Chandler, Arizona, with its reliably sunny weather, calm traffic, and meticulously mapped roads.

Will it soon feel normal to say, "Alexa, microwave one bag of popcorn"? Like a rebooted Sharper Image catalogue, Amazon is adding its talking artificial intelligence to a microwave, a wall clock, a wall plug, cars and more. The new gadgets all hook into the Internet, take voice commands -- and make the online retail giant even more central to home life. The question is: Will families see these connected devices as conveniences, new complications -- or spies? Amazon's goal is to assert leadership over Google and Apple in the still-nascent market for smart-home tech, with everyday appliances connecting to the Internet to automate operations -- and gather all sorts of data on our lives.

How much safer, smoother, and more efficient could driving be if cars could communicate with traffic lights while approaching an intersection, get alerted to jaywalking pedestrians, or talk to each other while roaring down the highway at 65 miles per hour? A peer-to-peer wireless technology called C-V2X can warn vehicles about obstacles that cameras and radars might not catch, connecting them to their surroundings in a way that could eventually help them drive themselves. Most of the demos involve people driving cars and trucks outfitted with special C-V2X chipsets and modems. The vehicles send and receive wireless signals 10 times per second and display certain types of information--such as warnings about oncoming pedestrians, storms, and accidents--as pop-up alerts on drivers' windshields or dashboards. The most recent C-V2X demonstration, which took place in Colorado on August 14, also connected participating vehicles to traffic lights, so drivers knew exactly when the lights would change colors.

A fleet of robotic jellyfish has been designed to monitor delicate ecosystems, including coral reefs. The underwater drones were invented by engineers at Florida Atlantic University and are driven by rings of hydraulic tentacles. The robots can squeeze through tight holes without causing damage. One expert praised the design but warned that the man-made jellyfish might be eaten by turtles. The flexible, 20cm-wide bots are modelled on the appearance of the moon jellyfish during its larval stage.

The work of an etiquette expert is never-ending. No sooner have you adjusted to a world in which the households you advise may have few or – whisper it – no staff, than the technology giants develop personal assistants using artificial intelligence. It is a whole new minefield and, as the Times reports, one already developing new expertise. One BBC tech executive told a conference audience on Tuesday that her solution to children developing poor manners due to Alexa, Siri and their rivals (the AI will respond whether you say "please" or not) was for adults in the house to say "please" and "thank you" to the AIs at all times. With that first step in mind, here is our extensive and scientific list of etiquette do's and don'ts when dealing with your AI assistant: Do: say please and thank you.