[From The Telegraph, where the story includes more images and two videos]
Her: Could you ever fall in love with a computer?
Spike Jonze’s Her, in which a man falls in love with his computer, was awarded Best Original Screenplay at last night’s Academy Awards. The concept may seem laughable, but advances in artificial intelligence are bringing us closer to our machines than ever before, says Rhiannon Williams
By Rhiannon Williams
03 Mar 2014
When I was a teenager, my friends and I used to hold conversations with a robot. Or to be precise, a chatterbot, a computer programme specifically designed to mimic human interaction through a series of exchanges. SmarterChild was widely available on instant messaging systems, and issued instant responses to whatever you asked it – generally queries about its sexuality and generic streams of abuse. While we were always aware we weren’t actually communicating with it, the novelty of appearing to ‘chat’ with a computer programme lead to over 30 million individuals adding SmarterChild as a contact on MSN messenger and AIM. In the 14 years since SmarterChild’s creation artificial intelligence has evolved exponentially, as has our attachment and reliance upon computers to run our lives. But could we ever actually develop feelings for them?
This is the premise explored by Spike Jonze’s Her, which won Best Original Screenplay at last night’s Academy Awards Ceremony, in which Theodore Twombly (played by Joaquin Phoenix) falls in love with his operating system Samantha, voiced by Scarlett Johansson. The genesis of Her was inspired by Cleverbot, a web application using a similar artificial intelligence algorithm to SmarterChild.
In an interview last year with The Guardian, Jonze described the first 20 seconds of interacting with one of these bots as “a real buzz”. “I’d say ‘Hey, hello,’ and it would say ‘Hey, how are you?’, and it was like whoa … this is trippy,” he said. “After 20 seconds, it quickly fell apart and you realised how it actually works, and it wasn’t that impressive. But it was still, for 20 seconds, really exciting. The more people that talked to it, the smarter it got.”
Lonely Theodore falls in love with virtual companion Samantha’s ability to speak to him like a real human through a small headphone-like device. “Every moment I’m evolving, just like you,” she purrs in his ear, as she rifles through his emails to get to the bottom of his relationship with his ex-wife and gently goads him into getting out of bed each morning.
On occasion I was acutely aware I was watching two of Hollywood’s finest flirting with each other – one present on screen, the other a disembodied voice. Tellingly Theodore’s ex-wife Catherine (Rooney Mara) is horrified by his confession he has fallen in love with his operating system, saying it made sense given that he couldn’t cope with the demands of a relationship with a human with needs.
But on the whole it’s an utterly absorbing love story which deconstructs the complexities of falling in love through the frame of technological innovation. Such is Samantha’s artificial intelligence, she longs to possess a physical body so she can walk around with Theodore and see the world as he does. I found myself able to suspend my disbelief he had developed feelings for a programme, given the tender nature of the pair’s interactions.
I asked Cloudera data scientist Sean Owen, and founder of machine learning company Myrrix, whether feeling compassion and even love for computer programmes won’t seem quite so weird in the future.
“Her is set in the near future; around 2050 or so. To some extent, our relationship with technology already matches that depicted in the film; for example, when you look around a subway carriage it’s not at all unusual to see the majority of passengers utterly engrossed in their phones. We already have that level of disconnect. But in terms of having a romantic relationship with our technology, we’re still quite some way off.”
Owen explains the evolution of artificial intelligence is an extremely complex journey which began in the tail end of the 1950s. “By the 1970s, programmes could be created that could answer series of factual questions, but they were extremely limited. Now the algorithms are much more sophisticated, and it’s much easier to feel you’re having an actual conversation with a programme such as the iPhone’s Siri as opposed to it firing answers back at you.”
The main barrier to developing an emotional attachment to our gadgets, he says, is that we’re still not entirely trusting of machines. “It’s more about whether people want that level of connection with their technology, which is tied into the concept of the Internet of Things. Our lives may become increasingly connected, but I’d say people still find the concept of their fridge ordering food for them a little creepy. Machines can help us to understand more about ourselves as dumb humans, and gain greater insight into why we behave the way we do. But I did find parts of the film quite disconcerting.”
Jackie Fenn, vice president and Gartner Fellow, says that many of Samantha’s capabilities, including speech, natural language recognition, and some conversational abilities already exist in current technology. “Once the computer can get smarter from new information, there’s nothing to stop it becoming as good as, and eventually better than, a person doing the same task,” she says. “So what’s to stop an OS from becoming a better companion than most humans? The more it interacts with you, the more it learns about what pleases you and what doesn’t, until it knows you better than you know yourself.”
One of Samantha’s most appealing aspects is her sense of humour (far from hampered by Johansson’s husky laugh). Expression of humour and creativity are the most challenging areas for artificial intelligence development, says Fenn, but that’s not to say it’s impossible.
“If a computer can learn what makes people laugh – and more importantly what makes you laugh – based on watching and analysing over time, there is no theoretical reason that a computer couldn’t eventually display and respond to humour. Similarly with music or art – by experimenting, analysing and learning, it could figure out which compositions create the best emotional resonance in the human brain.”
So it’s not inconceivable that a computer will soon be able to learn and deploy intelligence and interest in topics tailored to its individual owner, which may inevitably trigger emotional responses from our humane sensibilities. Dutch scientists found participants hesitated in switching off a robot cat that begs for mercy, despite full knowledge it was an android. If they perceived the robot as intelligent and agreeable, it took them three times longer to decide.
Theodore’s ex-wife Catherine tells him it’s apt he’s fallen in love with a computer, given his failures in his relationship with her as a human with needs. It’s worth bearing in mind the relative ease of engaging with a programme that doesn’t require the emotional maintenance in the same way romantic human relationships do, and how that may seem an increasingly attractive prospect to the time-poor or downright lonely individual.
Computers can, Fenn points out, already track our vital signs and establish how they change based on a person’s activities or sensory stimuli. “Put that together with the advances in brain-computer interfaces that determine intent and emotion directly from brain signals, and your OS will be able to figure out your needs without the need for a conversation,” she says. “Right now, much of the focus is on reading brain signals, but technologies such as transcranial stimulation have the potential to change brain states as well. If you wanted it to, your OS would be able to put you in a more focused or cheerful state of mind if it noticed you getting too distracted or grumpy.”
For all we know if a computer could effectively alter your state of mind, it may be less a question of whether we could fall in love with machines and more of whether we’ll have the capacity to avoid it. Fenn believes that while computers may be able to answer a question faster and more accurately than any person, it’s still going to be humans who decide what the right question to ask is. The Academy’s decision to award Jonze for his original foresight is an acknowledgement of what could very well be our future. One day.
Leave a Reply