Testing Turing

On the 60^th anniversary of Turing’s death, Saturday 7 June 2014, a computer passed the Turing Test. The computer, or Eugene Goostman as he’s known, is designed to simulate a 13 year old boy, and managed to convince one third of the judges that he was human – exceeding the 30% threshold that the Test sets. So what’s it like to judge the Test? There’s a fascinating account of one of the judges here. However, once the excitement of the breakthrough passed, researchers were left with some awkward questions. Is cybercrime going to be far more of a problem than it already is? What does it mean for a robot to have the functioning capacity of a human – should they have their own distinct legal personality? Some legal professors have argued robots must have legal personalities, following the Hobby Lobby case about corporations. The original paper by Turing is online and essential reading for those interested in Computer Science and robotics. As the word ‘robot’ comes from the Czech word for serf labour, we all need to be thinking about the repercussions of this story – whether you’re interested in Law, PPE, Philosophy or the sciences. 

What do you do if a substantial proportion of your population lack birth certificates, let alone passports? Since 2009, India has used AADHAAR, a massive biometric database – in fact, it’s the largest in the world. But it’s not been without its problems. Not least thanks to the fact that the system is not exactly fool-proof, with some Indians registering their dogs’ faces, but on a far more concerning level, India has no data protection laws yet, although the Supreme Court is doing all it can to protect citizens’ data. However, following the gripping Indian general election, the largest-ever election in the world to date and one in which holographic appearances were widely used, the future of the scheme is under threat. Although the scheme itself is a fascinating topic for Computer Scientists, Statisticians and Mathematicians, the role of the scheme and the Indian political scene as a whole is a brilliant case study for future Politics and Social Science students. 

Massive funding is rolling in for metamaterials – something if you don’t know about now, you might well do in the next decade or two. Last year, nanoimprinting raised the possibility of mass-manufactured invisibility cloaks (yep, Physics is cool). However, only last week, scientists in Singapore made a breakthrough in fabrication by increasing the thickness of fishnet metamaterials – whilst over in America a dash of magnetron sputtering helped produce one of the first metal-dialectric epitaxial superlattices (try saying that after a pint or two – to anyone who’ll listen…). So what’s all the fuss about? Well, metamaterials have tiny patterns on their surfaces which give them phenomenally exciting nanophotonic properties – basically, the ability to control light – which could be used to make anything from quantum computers to an optical microscope capable of seeing DNA, as well as fabled invisibility cloaks. It’s no surprise, then, that the metamaterials market might be worth around $643m by 2025, so look them up if you want to study Physics, Materials Science, or Natural Sciences (P).

TED talks, sensationalist journalism and the internet have all given the social sciences a boost over the last few decades. But are they facing a crisis of statistical analysis? Brave scholars have occasionally set out prank papers to prove their point – famously, in 2011, a highly respected journal published a paper by psychologists who had “proved” that listening to a Beatles song made people younger. Of course, mistakes are often made, and this website, Retraction Watch, is wonderful for procrastination until you remember that somewhere, someone is crying as years of work crumble to dust all thanks to a dodgy p value. This article has gathered together all of the hoaxes and explained the problems with the stats – and any student of Politics, Experimental Psychology, Economics, Mathematics and Statistics should have a good read.

Dark matter has appeared in the headlines this week – and just as more evidence of its existence has been found, more and more scientists are arguing that it simply doesn’t exist. So what’s going on? The famous Bullet cluster shows effects only possible (under current laws of Physics) had dark matter been not only in existence but travelling faster than the visible gas, and now scientists believe it is far from unique. Yes – you noticed – “under current laws of Physics”. That’s why some top theoretical Physicists have proposed adopting modified Newtonian dynamics instead, or “MOND”. Excitingly, this is becoming a much more feasible alternative, as MOND has given us correct predictions for speeds in Andromeda and galaxy-galaxy lensing. This is a topic to challenge even the best Physics and Mathematics applicants, and, from a modelling perspective, all you Computer Scientists.

The news that an Oxford spin-out firm is to start making “bucky-balls” is one of Oxford’s big stories this week. The carbon nanomaterial has unique qualities that could be used in everything from solar cells to MRI scanners. Existing issues with fullerines in solar cells have already led a Belgian group of scientists to make some exciting fullerene-free organic photovoltaic cells, which have just hit a record efficiency. “Carbon onions” have also been the subject of news in the nanotech world, as symmetry appears to be far more useful in mathematical modelling of nanostructures such as viruses than previously thought. All gripping stuff for the Mathematicians, Physicists and Natural Scientists out there.