Two years ago, I met Eugene Goostman, a guinea pig-owning, 13-year-old boy living in Odessa, Ukraine. Now this quirky character – in fact a software chatbot – is making headlines with the claim that on Saturday, he became the first piece of software to , the most famous test of machine intelligence.

Eugene, created by Vladimir Veselov, who was born in Russia and now lives in the US, and Ukrainian-born Eugene Demchenko who now lives in Russia, is certainly a clever, not to mention funny, piece of software. And it did fool 33 per cent of the people he chatted with into thinking Eugene was human. But here are a few reasons why the result may not be the milestone it seems to be.

Lies, damned lies and statistics

First conceived by the legendary Alan Turing in the early 1950s, the test challenges human judges to converse via a text interface with both hidden bots and humans – and say in each case whether they are chatting to a human or machine. Turing said that a machine that fooled humans into thinking it was human 30 per cent of the time would have beaten the test.

When I met Eugene, I was acting as a judge in a Turing test at Bletchley Park in Milton Keynes, UK, where Turing helped to crack the Nazi Enigma code, that was organised to mark the centenary of Turing’s birth. We had to sit at computer terminals and chat with invisible chatbots and humans via an instant-messenger-like program. With 150 separate conversations, or tests, carried out by 30 human judges, it was billed as the “most statistically significant” Turing test tournament yet to take place. In that tournament, Eugene fooled judges 29 per cent of the time – just 1 per cent shy of the 30 per cent needed to pass. In the most recent tournament, at the Royal Society in London, which involved the same number of tests and judges, Eugene upped his score by 4 percentage points. Sounds sort-of impressive, though with just 30 judges, only a handful need to be fooled to produce the improvement.

Other bots have “passed” the Turing test

How exactly to translate Turing’s ideas, conceived long before chatbots existed, into a meaningful test of artificial intelligence has been a matter of debate for some time. But as early as 1991, a bot called PC Therapist created by Joseph Weintraub took part in a Turing test and fooled 5 out of 10 judges – a pass rate of 50 per cent. Meanwhile, much more recently, in 2011, Rollo Carpenter’s Cleverbot chatted with 30 humans in front of a live audience of over 1000 and fooled 59.3 per cent of the judges and audience into thinking it was human.

In an email I received inviting me take part in the Turing test this year, one of the organisers, Huma Shah at the University of Reading, UK, stated that because of these previous attempts, “passing” the Turing test was not the point of the latest tournament:

“The purpose of this June’s Royal Society experiment is not to find if a machine can now pass one interpretation of Turing test success, 30 per cent, especially as a machine surpassed this with a deception rate of 50 per cent back in 1991 and machines do not have anywhere near the conversational skills of HAL from 2001: A Space Odyssey.”

Rather, the organisers wanted to “encourage/inspire children to take up computing and robotics, is also about awareness and prevention of cybercrime”. Worthy goals – but this isn’t really about Skynet.

Eugene is only 13 years old

Would you consider a teenager the pinnacle of human intelligence? Probably not. This gets at something that is both clever – but also limiting – about Eugene. Passing the Turing test is touted as being the moment that machines have got us, since they can perfectly imitate people. But it doesn’t address which people the machine must imitate in order to pass, raising the question of whether a 13-year-old boy really counts. When I met Veselov in 2012, he said: “Thirteen years old is not too old to know everything and not too young to know nothing.” In other words, it’s old enough to hold a conversation, but much easier to mimic, than, say, Richard Dawkins, or Stephen Hawking. By aiming low, Eugene succeeds at being a realistic human character – but is he really all that smart?

Intelligence is about a lot more than chatting

The Turing test has come to symbolise machine intelligence but it only tests machines on their ability to chat – and people are capable of so much more. So in recent years, there have been calls to come up with an upgrade that would compare computers . In the Visual Turing test, for example, computers are challenged to mimic our visuo-spatial capabilities.

Others worry that it’s wrong to use ourselves as the benchmark of intelligence – and that this may lead us to miss out on the discovery of a true mega-brain. AIs – or intelligent aliens, say – that solve problems in very different ways may be as smart as us, or even smarter, but would fail a Turing test merely because of its human-centric bias. With this in mind, efforts are underway to create a universal intelligence test that measures smarts via objective qualities like the ability to compress information and make decisions.

None of this should take away from the achievements of Veselov and Demchenko – but it does help to put their creation in context. And just to add: when I took part in the Turing test, I correctly classified all bots as such – including the quirky Eugene.

Video: Pitch drops again in world’s longest experiment

The pitch has dropped – again. This time, the glimpse of a falling blob of tar, also called pitch, represents the first result for the world’s longest-running experiment.

Sadly however, the glimpse comes too late for a former custodian, who watched over the experiment for more than half a century and died a year ago.

Up-and-running since 1930, the experiment is based at the University of Queensland in Australia and seeks to capture blobs of pitch as they drip down, agonisingly slowly, from their parent bulk.

It was pipped to the post last year when a similar experiment, set up in 1944 at Trinity College Dublin in Ireland, captured the first ever video footage of a blob of pitch dropping.

In that instance, the blob separated from its parent bulk. By contrast, the Australian team filmed the collision between the ninth blob ever to fall and the eighth blob, which was sitting at the bottom of their beaker – but the ninth blob is still attached to the pitch above it.

Still, the Australian result is important because the experiment has a better set-up, says , a member of the Trinity College Dublin team who used those results to calculate the pitch’s viscosity. “Theirs is in a glass container; they measure the temperature, measure the humidity as well,” he says. “Ours, we don’t really call it an experiment. It was really just sitting there on a shelf, going back to the 1940s.”

Near miss

The fact that both experiments dropped within a year of each other is “just pure luck”, says Hutzler. Hot summer weather in Ireland last year may have influenced the timing.

The Queensland experiment already and won an IgNobel prize in 2005. It was set up by physicist Thomas Parnell to illustrate that although pitch appears solid, shattering when hit with a hammer at room temperature, it is actually a very viscous liquid.

The eventual result follows several near misses, according to the University of Queensland. John Mainstone, who oversaw the experiment for more than 50 years , missed observing the drops fall three times – by a day in 1977, by just five minutes in 1988 and, perhaps most annoying, in 2000, when the webcam that was recording it was hit by a 20-minute power outage.

“It’s a pity of course that the person in charge died about a year ago, so he never saw the drop,” Hutzler says. “He would have enjoyed that.”

Honey flow

The eighth and ninth drops each took about 13 years to fall, says current custodian Andrew White. By contrast, the seven drops that fell between 1930 and 1988 did so faster – at an average rate of one drop every eight years.

The next step is to see how long it takes the ninth drop to separate from the pitch above it: “It may tip over quickly or it might slow right down and take years to break away,” says White.

You can keep an eye on the ninth drop’s movements via a . The University of Queensland says it will work out who was watching when the pitch dropped and record their names for posterity.

The drop experiments show that the physics of a drop forming in a viscous material is still not well understood, Hutzler says – although he doesn’t think watching pitch for decades is necessarily the best way to study it. Using honey or some other less viscous fluid would give you better statistics.

“I think these experiments capture the imagination just because they go on for such a long time,” he says. The video of the drop in Dublin quickly went viral on YouTube. “Ironically, you have a very slow event happening, but the news spreads very quickly.”

First there was Bert and Ernie. Now there is Big Bird. This isn’t a flashback to the birth of , but the latest in the hunt for subatomic particles from deep space. Big Bird is the playful nickname given to the highest-energy cosmic neutrino ever detected.

Neutrinos are nearly massless particles that are churned out by the sun and by processes on Earth such as radioactive decay. They are also generated in deep space by objects such as supernovas and the supermassive black holes at the centres of galaxies. Only recently have we started systematically detecting these cosmic neutrinos, which tend to hit higher energies than those produced in our solar system.

Astronomers hope to read them as if they were cosmic messages. Many particles get deflected or absorbed as they cross the universe and so don’t hold a faithful record of where they came from. But neutrinos barely interact with anything and so arrive at Earth almost unscathed, which means they could be used to piece together details of the distant events that produced them.

Bert and Ernie

Last year, the IceCube neutrino detector at the South Pole, designed specifically to investigate cosmic neutrinos, reported its first catch – a signal consistent with two high-energy particles, which the team nicknamed Bert and Ernie. Each had an energy of about 1 million billion electronvolts, otherwise known as a petaelectronvolt (PeV).

These were soon followed by an additional haul of 26 neutrino candidates, not quite as energetic as Bert and Ernie but still most likely cosmic in origin. Then in November, all .

By then, the team had also peeked at some partially analysed data and caught a glimpse of what looked like a neutrino of even higher energy, which they aptly nicknamed Big Bird.

Now Big Bird has official status, appearing in two reports presented last week by IceCube team members at the in Savannah, Georgia. “It is more energetic than Bert and Ernie, by approximately a factor of two,” says IceCube member Chris Weaver, “and so it is the most energetic neutrino ever detected.” The updated analysis also confirmed that we have not yet seen any neutrinos of higher energy than Big Bird.

Clusters wanted

But though Big Bird a record-holder, it’s not a revolutionary. The fact that its energy is twice as high – rather than 10 times higher, say – as the energies carried by Bert and Ernie reinforces a suspicion that there is a peak in neutrinos at a few PeVs. Some researchers attribute this suspected energy distribution to mysterious dark matter; others think that there simply aren’t cosmic sources for neutrinos of higher energy.

The next major goal, says Weaver, is to start figuring out which objects produced these neutrinos, most likely by finding several cosmic neutrinos streaming from the same point in the sky. “Seeing a cluster of events on the sky could help us better pin down where the source is actually located,” he says. “In the long term, having more neutrinos would be much more interesting as it would allow us to actually study the energy spectrum of the source and learn about the processes which produce the neutrinos.”

Video: Big Bang breakthrough explained with bath towel

Feel like you need a giant IQ to understand Monday’s landmark announcement about gravitational waves from the birth of the universe? We don’t think so: our video (above) explains the breakthrough using nothing but a towel, an apple and a ping-pong ball.

On 17 March, physicists working on the , announced that they had glimpsed gravitational waves – ripples in the fabric of spacetime – dating back to the universe’s birth.

The observation, via tell-tale swirls in maps of relic light from the big bang, represent the first clear detection of gravitational waves, which were first predicted by Albert Einstein.

The discovery also allows us to peer back in time further than anyone thought possible. That could have all kinds of knock-on effects for our understanding of the universe, perhaps even helping to shore up one of the most exciting, and baffling, ideas in cosmology: the multiverse.

The idea that our universe is just one of many is a whole other story, but to get handle on how we got to this point, watch the video.

Colin Angle is co-founder and CEO of iRobot in Bedford, Massachusetts. The company recently created a , and this week released a telepresent robot called Ava 500. Its Packbots will also look for bombs at the football World Cup in Brazil this year. We caught up with Angle earlier this month

People dress up Roomba vacuum cleaners. Soldiers . Are these responses to your robots a surprise?
If you ask someone who doesn’t own a Roomba if they would name it, they almost violently say, “No, why would I name my vacuum cleaner?” Yet once they own one, over 80 per cent of people do. In the beginning of Roomba, we all took turns answering the support line. Once, a woman called and explained that her robot had a defective motor. I said, “Send it back, we’ll send you a new one.” She said, “No, I’m not sending you Rosie.”

Do you sympathise with that personally?
Yes. My Roomba’s name is Roswell. There is the moment when you are sitting on the couch and Roomba turns itself on and goes out and starts working. You really appreciate it because it works hard for you, and it deserves some kind of recognition.

How else are robots changing us?
We will have a generation who grow up assuming robots in their world in a way you and I never did. When my daughter was 3, she was eating Cheerios and spilled some on the table. So she swiped them onto the floor. I said, “Darcy, what are you doing?” She said: “Don’t worry, Daddy, the robot will get it.” I didn’t know whether to be horrified or proud. It was this idea that homes take care of themselves and robots are part of that.

Your bomb-disposal robots will be at the football World Cup in Rio later this year. How will they be used?
At the World Cup, there is a constant risk that you might find a bag or some object that has been left behind and no one is quite sure what it is. To bring in a full bomb-disposal team for each item can be very time-consuming. The PackBot can go over rough terrain, climb stairs, pick things up and also be operated from a safe distance.

This week you released a telepresent robot that allows someone to send a robot to a meeting instead of attending in person. How is that different to other telepresent robots, such as the ?
The goal with the is a system that allows you to project yourself with comparable charisma to being there in person. A big display and high-definition video convey the social cues on the person’s face. Ava also stands up and sits down; if you were seated but I was standing up, it would be a very short interview because you would feel uncomfortable. One of the biggest differences with the Ava 500 is navigation.

How does the robot get around differently?
Other telepresent robots are remotely driven, basically with a joystick. But in a business setting, you may not know where to go. Worse, most buildings don’t have continuous Wi-Fi. If you hit a dead zone, you are a paperweight. With our interface, the app will say, where would you like to go? The robot then uses a technology we have developed to quickly build maps and locate itself within them. Then it drives there autonomously.

As robots get more skilled, at what point should people start to be scared of them?
A robot take-over is not going to happen. Robots are going to help us with various aspects of our lives, but what will be weirder is the integration of robot technology into ourselves. Today if you lose your hearing, you can get an implant that restores it. The strange part will be when you can get even better hearing. Augmenting our bodies with technology will have far-ranging ramifications and is going to happen far sooner than humanoid robots coming in to take my job – or to take over.

(Images: Kacper Kowalski/Panos Pictures)

GAZING at these winter landscapes, do you see lines carved by nature or humans? It’s hard to tell – which is exactly what aerial photographer wants.

“I love to see this border between nature and human activity,” says Kowalski, who took the shots in Pomerania, north-west Poland, where he lives. To capture some images, he dangled from his paraglider. For others, he used – a small, open-air helicopter that resembles a flying motorcycle. Such detail wouldn’t be possible from an ordinary plane because the windows would get in the way.

It isn’t easy to operate a camera while braving the cold and wind, particularly in the paraglider. “The most uncomfortable thing is that I have to have my hands on the brakes for control, which are above my heart, so blood doesn’t flow to my hands,” he says. But such hurdles help to make his images unique: “Even pilots don’t often see this winter landscape.”

The image at the very top shows a stream snaking through a snow-covered field dotted with trees. It’s in contrast to the image just above: Kowalski thinks this thin black ribbon, at the edge of a lake, is human made. Bordered by what look like piles of snow, it is probably a path dug by ice fishers to make crossing the ice easier. In the image below, the black lines are once again natural, the result of water seeping up through splits in the ice on a lake.

The uncertainty over what is natural fits with Kowalski’s desire to fire our imaginations. “People see so many things, and it is so hard to surprise them,” he says. “I like not having too much information about the situation because then the image is working for longer.”

This article appeared in print under the headline “Pure imagination”

Read more: “2014 preview: 10 ideas that will matter next year”

From time-travel movies to wormhole entanglement, 2013 delivered a load of mind-boggling ideas. It was also the year that several big experiments delivered the goods we were hoping for: the Planck telescope provided the most accurate map of the big bang’s afterglow; the discovery of the Higgs boson was confirmed, winning those who first predicted it the Nobel prize; and neutrinos from outer space were finally glimpsed. But as attempts to solve a knotty problem known as the black hole paradox reveal, we’re in no danger of running out of mysteries quite yet.

First real time-travel movies are loopers
Hollywood has played with time travel for decades, but now physicists have captured on camera what travelling to the past actually looks like.

Planck map reveals birth, life and death of a cosmos
The best map yet of the cosmic microwave background – the big bang’s “afterglow” – provided both novel insights and strange anomalies.

Neutrinos from outer space open new eye in the sky
Supermassive black holes, giant exploding stars and dark matter may give up their secrets now that a telescope buried under the Antarctic ice has detected neutrinos from deep space.

First nanotube computer could spark carbon revolution
Watch out, silicon: a functioning computer was built from carbon nanotubes – and it comes with its own operating system and software.

Pitch drop caught on camera after 69-year wait
One of the world’s longest-running experiments climaxed when a finger-sized bulb of pitch (bitumen) separated from its parent bulk and dropped into a beaker. For the first time ever, the fleeting event was recorded on video.

Google and NASA team up to use quantum computer
Big-name clients for quantum-computer maker D-Wave signal that the devices are going mainstream, and may even power the wearable computer Google Glass, due to be released to the public in 2014. But though the computers are fast, it’s still not clear whether they are truly quantum

Wormhole entanglement solves black hole paradox
Tunnels through space-time and spooky action at a distance – two of the most baffling ideas in physics – may be different manifestations of the same thing. It’s an insight that could pave the way for a theory of quantum gravity, and solve a niggling paradox surrounding black holes

Elusive Higgs wins physics Nobel, shared with Englert
In 2012, they found the particle. This year, it earned its stripes. The 2013 Nobel prize in physics went to Peter Higgs and François Englert for developing the theory of how particles acquire mass via the Higgs boson and accompanying Higgs field. The prize followed the particle’s official recognition as a Higgs boson earlier this year.

First fluid knots created in the lab
Knots are already a subject of fascination to mathematicians – these new beauties are made of water and could boost our grasp of aircraft wings and weird quantum superfluids.

New 17-million-digit monster is largest known prime
A distributed computing project found a prime number larger than any other known. 2013 also saw serious progress on an intractable problem known as the twin prime conjecture.

(Image: HAP/Quirky China News/Rex)

Yutu’s journey has begun. The Chinese moon rover has moved away from its parent spacecraft, Chang’e-3, which successfully landed on the moon on Saturday – and the pair have already started snapping photos of each other and beaming them back to Earth.

Chang’e-3 took the image above, which shows the rover’s tyre tracks in moon dust after it was deployed down a ramp from the lander. Yutu – whose name means Jade Rabbit – was also videoed taking its .

Tiny liquid volcanoes that spray beams of charged particles could make space history next year.

They are one of two technologies vying to be the first to let cheap, miniature satellites called CubeSats fly in formation, switch orbits or voyage to other planets – feats usually reserved for large, expensive craft. They could even provide us with a global Wi-Fi system on Earth.

leads a team working on CubeSat propulsion at the Massachusetts Institute of Technology. He has been given the go-ahead to launch two propelled CubeSats in 2014 – one funded by the US Department of Defense, the other by private donors.

Meanwhileat the University of Michigan in Ann Arbor, who leads a rival project, announced that his team also has private funding – and a slot to launch their CubeSat on a NASA rocket next year. The race is on.

CubeSats are made from off-the-shelf components. The initial aim was to make access to space easier and more affordable.

Each one weighs just 1 kilogram, has 10-centimetre-long edges and can hold components like sensors and cameras. They are typically put into low Earth orbit by a rocket, where they remain for around 6 months, before spiralling in and burning up in Earth’s atmosphere. Although they have made space accessible to groups who wouldn’t otherwise have been able to afford it – – CubeSats haven’t done much cutting-edge science. “They were considered like toys,” says Lozano.

Some CubeSats have basic steering, but getting them to change orbits, let alone visit other planetary bodies, requires new technology. That’s because even today’s most efficient propulsion method, the ion engine, doesn’t scale down to CubeSat size.

Instead, Lozano and his colleagues will propel their craft with an unusual substance called an ionic liquid, made solely of positively or negatively charged ions. In the engine, a reservoir of ionic liquid soaks into a porous, metal chip and forms tiny pools in the pores of spikes on its surface. When a small electric field is applied, these pools morph into cones, which amplify the electric field so that it is strong enough to pull away ions in a steady beam (see below).

The process is self-sustaining. Fresh liquid gets sucked onto the chip when ions are emitted, just as tree roots suck in water when vapour escapes the leaves. The result is an array of between 500 and 5000 focussed ion beams that stream from each of the eight chips on the CubeSat when the electric field – the strength of which acts as the engine’s throttle – is applied.

Lozano’s team have fired the thrusters in the lab and calculate that just 8 grams of ionic liquid will propel a 2 kilogram CubeSat and change its orbit by 100 kilometres. The team plans to test this in one of its launches next year. Eventually, the aim is to send such a satellite to an asteroid to collect a scoop of dust.

They have stiff competition. Longmier’s team . Their CubeSat , or CAT, uses xenon, like traditional ion engines. The difference is that the CAT adds an intense and carefully shaped magnetic field that stops xenon ions from hitting the engine walls and going to waste. Using a permanent magnet means that no power is required to produce the field. They say a future version could also use water instead of xenon, which would make it easier to refuel.

Longmier’s team began their first crowdfunding campaign on the Kickstarter website in July. Although they failed to raise their $200,000 target, the appeal prompted a private donor to offer to pay for the technology and a launch next year aboard a NASA rocket. They hope their propelled CubeSats will one day fly to Saturn’s moon Enceladus and Jupiter’s moon Europa, both of which hold water – and perhaps life. A fleet of CubeSats with propulsion in orbit around a planet or moon can do a lot of things that big expensive satellites cannot, such as monitoring several locations in the atmosphere at once.

Propelled CubeSats could even be useful back here on Earth. Creating a universal “satellite Wi-Fi”, like existing satellite phone coverage, would require thousands of big satellites, which is prohibitively expensive. But you could dump a thousand CubeSats in one place then spread them out to the right points, for a fraction of the price.

Longmier’s team has , which could fund some add-ons, including a camera. No matter what happens, the team already has enough money to launch and propel the CubeSat next year. “We might have a little space race on our hands,” says Longmier.