Another great success for Lipstick Robot
The queen of bad robots lives in Sweden.
It’s not a title she chose for herself, Simone Giertz assures me. It’s something that people started calling her ever since she began sharing her hopeless creations with the internet seven months ago.
You might have seen them – the robot arm that smears lipstick on and around her mouth, or the “wake-up machine” that enthusiastically slaps her in the face with a rubber hand. One nightmarish invention is supposed to chop vegetables using two mechanised knives on metal springs. It’s so aggressive she has to feed the tomatoes in with tongs.
The first Giertz robot I saw was the Breakfast Machine. In her 30-second demonstration of the machine, it manages to go awry in every way possible. First, an expensive robo-arm with a suction cup for a hand grabs a box of Cheerios, pours them in the vague direction of a cereal bowl, and tosses the box onto the floor. Then it sloshes milk in with about as much success. Finally, it picks up a plastic spoon and delicately attempts to scoop up a bit of cereal, getting exactly nothing in the process. Giertz sits blithely by throughout, reading a book and occasionally leaning in to take a mouthful.
Behold the Breakfast Machine
I was enthralled. It’s hilarious, obviously, but also familiar in an off-kilter way: it’s the Silicon Valley start-up that dreams of optimising our morning routines; it’s the academic project that throws grant money at training robots to make simple meals.
Giertz harbours no such lofty ambitions. “I never set out to build something actually useful,” she says. “Almost all of my projects start out with a thing I would want to automate in my life, and then I just go with the over-the-top solution.” Brushing your teeth, for example, is boring. A helmet that does it for you? Sure, why not.
Giertz learned how to make her contraptions by tinkering with an Arduino starter kit. Eventually, her videos attracted the attentions of the Reddit community, where she is now a fixture on the “shittyrobots” board, holding the all-time most popular post.
So what qualifies as a good “shitty robot”? Adorable robots are not allowed, nor are non-physical robots such as. Useless or suicidal robots – like this that kills itself by drinking Coke – are their own special subtype.
A true shitty robot is one that is out in our world, doing a real job, and totally backfiring. Unlike Giertz’s Breakfast Machine, many of the terrible robots on display here are accidents. They weren’t built to entertain, but to get stuff done. Their falls, spasms, mistakes and self-inflicted damage occur in the pursuit of true artificial intelligence.
A kitchen robot tries to flip a pancake and instead sends it flying onto the floor. A soccer robot at theshuffles and shuffles in search of the perfect spot to kick the ball, then misses entirely and falls flat on its face. Perhaps the classic of the genre is that picks up a bin, makes an empty gesture of emptying it, and then calmly spews trash all over the road.
Videos like these are undeniably silly. They fit right in with the internet’s rich tradition of celebrating epic fails. Their popularity is unsurprising. But some think there might be a deeper reason why we find these videos so satisfying.
Our feelings about robots are complicated. Many – including me, in the pages of New Scientist – have described the promise of the coming robot age in lush detail. Robots will, , arrange our schedules and . In short, they will be the perfect servants.
At the same time, every advance in artificial intelligence has people breaking out the musty jokes about our future robot overlords –.
“There’s a whole trope out there about this kind of existentialist threat,” says, who studies robotics at Carnegie Mellon University in Pittsburgh, Pennsylvania. “It’s been whirling about in the Jungian pysche of humanity anyway, because of movies like Terminator,” she says. But warnings by the likes of Elon Musk and Stephen Hawking, she says, “are really making it start to get into people’s brains”.
Nourbakhsh is referring to an open letter circulated last year, in which tech luminaries warned scientists to start paying attention to the, including, but not limited to, the sudden disappearance of jobs, automated killing, and even the loss of “meaningful human control”. – showcased by this month’s – keep these fears simmering in our collective imagination, he says.
Seeing a robot struggle with something simple soothes a little of that anxiety. Oh, you can’t even open a door? Good luck taking over the world, puny machine.
Watch this Darpa robot fall hard
Humans build machines to help us manage the world, but the technology can in turn create problems. Mark Coeckelbergh, a philosopher at the University of Twente in the Netherlands, calls this shifting relationship the “anthropology of vulnerability”.
“Laughing at a clumsy and failing robot is also a means to cope with a serious problem,” he says – the problems that Musk or Hawking are talking about, but also just the mundane menaces of everyday life. Sometimes our dependence on machines just makes our lives harder, like when the printer jams right before the big report is due, or when the calendar that was supposed to automate your schedule inexplicably drops a couple of meetings.
Giertz’s inventions hit that vulnerable spot with unerring accuracy. “It does make you feel good to see a robot fail at something you do every day,” she says.
More on these topics: