Download the FREE SBS Radio App for a better listening experience
Already robots are taking our jobs and computers may soon be smarter than humans in every possible way.
Could it get worse from here or will robots make our lives better?
Signe Dean reports.
[two voices talking in movie excerpt] "Open the pod bay doors, HAL." "I'm sorry, Dave. I'm afraid I can't do that." "What's the problem?" "I think you know what the problem is just as well as I do."
So unravels an iconic scene from Stanley Kubrick's film 2001: A Space Odyssey, in which HAL 9000, a sentient (or feeling, perceiving) artificial intelligence computer turns on humans.
It is the stuff of science fiction - for now.
But with A-I and robots already present in everyday life, it may just be a matter of time.
Toby Walsh is a Professor of Artificial Intelligence at the University of New South Wales and also at NICTA, Australia's Information and Communications Technology Research Centre of Excellence.
He says many of us already use artificial intelligence every day.
"Anyone using a smartphone who is using Siri or the Microsoft equivalent, and asking questions to your smartphone, is using a bit of artificial intelligence. There's a very smart program there that is understanding your speech and then trying to answer your question in an intelligent way."
Banks use AI systems to track whether a person's credit card is used in a suspect way.
Artificial intelligence also assists pilots who fly planes.
Even driverless cars are currently being tested on roads.
Now, these AIs are only good for one or a few tasks, while a human can do much more.
But some experts argue it won't be long before we really do have super intelligent computers - like HAL.
Professor Nick Bostrom is a philosopher at the University of Oxford, and author of the best-selling book Superintelligence.
He explains what a superintelligent AI is.
"Basically any AI that would dramatically outperform humanity in all practical and relevant fields. I'm not focusing on just domain-specific AIs, AIs that are just better at playing chess, or just better at counting numbers, but AIs that have the same general purpose, smartness, the same powerful learning ability and planning ability that human brains have."
For now, such systems don't exist.
But with AI research going in leaps and bounds, it may be just a matter of time.
"It's true that we don't have any such systems today; I think though that they will be developed eventually, assuming that science and technology continue to progress. But that there's great uncertainty as to the timescale for this. It's been a goal since the beginning of artificial intelligence as the field to create thinking machines and researchers have been plugging along on this problem for a number of decades, and I think they will continue to do so until finally the goal is reached."
It's pretty much impossible to predict how soon we will have superintelligent computers.
Professor Toby Walsh explains that the typical time frame is one of roughly forty years.
"People have been predicting superhuman intelligence for computers for some time now, and if you ask the experts in the field they still say it's 30 or 40 years away. So we still have some time before we have to think about these issues."
Professor Bostrom agrees that it's a matter of probability, not an exact date.
"I think some number of decades seems like a vague, but plausible statement to make. It could happen in the next century, but I think there's a real chance it could happen within the lifetime of a lot of people alive today and listening to this program."
A superintelligent machine would have the processing power to solve a lot of humanity's problems.
Professor Walsh explains why embracing A-I research is valuable.
"There are certainly a huge number of problems that face society today; global warming, increasing population, increasing inequality in the society, the global financial crisis. All of these things that are stressing the planet could be tackled if we actually made better decisions, if we had computers to help us make better decisions."
However, superintelligence is also super-powerful.
And that can be dangerous, says Professor Bostrom.
"I think there is significant existential risk associated with this transition to the machine intelligence era. There are threats that could jeopardise the survival of our species, our human values. Intelligence is a very powerful thing, and something that was far more intelligent than we humans are, potentially could be a great threat to humanity, unless it is very carefully designed so as to be aligned with our values, so it has to be on our side."
If a machine is more powerful than a human in every possible way, why should it take orders from us?
Instead, it could decide that humans are harming the planet.
If we're lucky, we may be kept as pets.
And Professor Bostrom thinks that controlling a superintelligent A-I will be tricky.
"I think it looks like it could be a very difficult technical challenge, how to control a superintelligence. And moreover a challenge which we might only get one go at. Once you have a superintelligence that is unfriendly and hostile, it might be too late, it might not be possible to put it back, to put the genie back in the bottle. So we maybe got only one shot to get this transition to the machine superintelligence era right."
For now we are still in charge of computers and also in charge of robots.
Professor Mary-Anne Williams is a roboticist at the University of Technology Sydney, where she runs the MagicLab.
She explains that robots are not what science fiction makes them out to be but they are a reality today.
"Essentially a robot is a computer that can make decisions using sensors and can undertake real, physical work. And that's what separates a robot from computers on the internet."
It's important to make sure that people and robots get along.
Professor Williams explains this is what the field of social robotics focuses on.
"In my lab we've decided to focus on social robotics, because it's the future - robots are entering society and the workplace increasingly, and we need smarter technologies to govern them, and to allow them to reach their full capability. So if they can make smarter decisions using the knowledge that they have and the data that they can collect, then that's gonna be a better outcome for not just industry, but also society."
There are many uses for robots today.
Specially programmed companion robots help autistic children with learning.
Robotic arms help surgeons perform tricky procedures.
In factories, robots increasingly displace assembly chain workers.
In fact, robots are downright taking away our jobs.
In June a major C-E-D-A report indicated that in the next ten to fifteen years almost 40 per cent of Australian jobs would disappear due to technology advancement.
Professor Toby Walsh from UNSW compares it to the industrial revolution.
He also explains that no jobs are safe.
"Any job you can think of can in some way be automated. It used to be we were thinking that it was the more manual jobs - building cars in factories were the jobs that the robots were gonna take over, now it's increasingly likely that some of the middle class jobs, the jobs that we thought that we could go to university and make ourselves safe are being automated. Lawyers, accountants, doctors and so on, their jobs will also be increasingly automated."
Even writers are not safe; chances are you have read financial reports or sports news compiled by a machine.
However, as the robot era continues, new professions will emerge, too.
Professor Williams explains that's not necessarily a bad thing.
"They will continue to push into the workplace, and increasingly we'll see them going up the value chain, displacing people. But they'll also be creating new kinds of work that we haven't even thought about, and they'll be creating new jobs for people. It is well known that technology advances are essentially good for society, they allow us to be more innovative, and they also allow people to do more interesting kinds of work. And people are very adaptable."
If you are choosing a future profession, things may not be so simple a few decades from now.
"I think it's something that we all need to think about, because we all need to think about the nature of our employment, the nature of our children's employment, what we should be doing to make ourselves future-proof."
We may use robots to vacuum the floor, leaving more time to spend with the kids.
But there is a dark side to this technology as well.
Automated machines such as robots are already used in warfare.
This makes some experts like Toby Walsh worried.
"Killer robots is an area where we have to worry about very quickly. There is already a campaign at the United Nations to introduce a moratorium, a ban on killer robots. I would be surprised if in the next two or three years there wasn't some sort of ban put in place, and I think that is needed, and that is very timely. Killer robots are gonna take many forms, drones is one of them, autonomous vehicles driving around with machine guns strapped to them is another, but also there's gonna be killer submarines, almost all forms of warfare will be touched by this."
Meanwhile Professor Mary-Anne Williams at UTS would argue that there are other things to be concerned about.
"Of course, killer robots are super-dangerous, I mean a robot walking around with a gun or any kind of weapon is very, very threatening, and also very, very real. In terms of perspective, I think there are much more serious concerns about advanced technologies, and many of them are around privacy. The NSA is far more dangerous than killer robots. I mean it is extraordinary that an organisation like the NSA would know more about Australian citizens than their own government."
From being enslaved by a superintelligent computer to being killed by robots, the future comes with certain risks.
But as we move from science fiction to science reality, there is every chance that in 40 years our society will be improved thanks to technology.
"Robotics is a very exciting area, because it provides a way for us to understand ourselves. Understand the way our brain works, the way we make decisions, and it also provides hints to a lot of the old philosophical problems around humans on the planet, and how we evolved, so we're finding all kinds of insights that allow us to understand ourselves better. That's what gets me very, very excited, and I think in the long run robots and advanced technologies will transform society and make it more free and more fair."