Welcome to the future - when machines will rule
Will AI be the death of us or usher in a brave new world? Farren Collins joins the fray
The warning signs have been there for years. Isaac Asimov even went as far as codifying them in his fictional laws of robotics, which declare: "A robot may not harm humanity, or, by inaction, allow humanity to come to harm."
It's a law we may soon need in real life, because the machines are undoubtedly coming. But are they out to get us? The fourth industrial revolution has arrived and it has come in the form of artificial intelligence. In our efforts to do less so that we can do more, we have developed self-learning technology that can even do our thinking for us, and it is proving to be better at it than we are.
More than a thousand years after we developed the first analogue calculator, the abacus, we are at the point where cars drive themselves and the smartest game-playing humans are beaten by a bot in a highly complex game that requires tactics like feinting and denial.
It took just two weeks of playing against itself for OpenAI to learn the battle video game Dota 2 and beat the game's best human players. That victory - earlier this month - came nearly two decades after IBM's AI, Deep Blue, beat chess grandmaster Garry Kasparov, and just months after Google's AlphaGo wiped the floor with 18-time world champion Lee Sedol in the boardgame Go, in a match watched by 200 million people.The company that developed OpenAI is owned by one of AI's foremost scaremongers, South African-born tech entrepreneur Elon Musk, who thinks it could bring about the third revolution in warfare. On Monday he and 116 other experts signed a letter to the UN, asking it to ban "killer robots" and halt the development of autonomous weapons.
"These can be weapons of terror, weapons that despots and terrorists use against innocent populations, and weapons hacked to behave in undesirable ways," the letter reads.
Human-operated drones are already being used in place of foot soldiers. Future capabilities for unmanned aerial vehicles could include autonomous takeoff and landing, so the apprehensions are not completely misplaced.
Earlier this year, Musk, who is also CEO of the world's first private space company, SpaceX, told US governors that AI was the "biggest risk we face as a civilisation"."Until people see robots go down the street killing people they don't know how to react," he said.
Musk recently had a public spat with fellow tech billionaire and Facebook owner Mark Zuckerberg after he said AI could be more dangerous than the nuclear powerhouse and casual missile-testing North Korea.
But there is arguably cause for suspicion when someone so critical of the technology is also leading advancements in AI research. You can almost imagine Musk wanting to qualify his statement by adding "unless it is in my hands". But that is only a suspicion.
Zuckerberg, on the other hand, is embracing the technology and last week launched a new AI for the Facebook Messenger application that makes suggestions and recommendations to users based on the conversation they are having.
It didn't take Zuckerberg long to hit out at Musk for his remarks, which he cheekily called "pretty irresponsible".
"I think people who are naysayers and try to drum up these doomsday scenarios - I just, I don't understand it," Zuckerberg was reported as saying.But Facebook itself abandoned an AI project last month after two chatbots it created to negotiate the trade of hats, balls and books went off cue and started communicating in their own bizarre language.
Essentially what the chatbots did was to create their own shorthand, which didn't make sense to humans and frankly was kind of creepy, because they weren't incentivised to communicate according to the rules of the English language.
In the end the experiment was canned.
The name-calling between two of the planet's most influential people continued on Twitter where Musk - who wants proactive regulation for AI - said he had spoken to Zuckerberg and determined that he had a "limited" understanding of the subject.
Regardless of which of them is right, it is clear that artificial intelligence is here and will be an increasingly important part of our future.
And while Musk was cautioning against a Skynet-run world with self-operating weapons ready to end us, the International Joint Conference on Artificial Intelligence, where 2000 AI and robotics experts met to discuss the technology's future, was quietly taking place in Melbourne, Australia.
But is Musk justified in his fear of a Terminator-type future? Or can Pandora's box be contained once it is opened?For the most part we appear to display a fair amount of cognitive dissonance when confronting the question of whether or not AI is good.
We want our phones to be able to talk us and we use smart technology everyday to make our lives more efficient, but the mention of AI still often has us drumming up horror images of dystopian societies where machines have replaced humans at the head of the pecking order.
Maybe that's just Hollywood exposing our gullibility. There can be little doubt that film and literary adaptations of AI have played a big role in informing our understanding of it.Our obsession with anthropomorphising robots as sentient beings is well documented and helps us ask ethical questions around the use of AI.
We have gone from depicting machines as enslavers, in films like The Matrix, to modern series such as Westworld, which is set in a theme park where humans live out fantasies with the robots that they build.
Ian Rijsdijk, from the Centre for Film and Media Studies at the University of Cape Town, described the pursuit of AI as Promethean, the "idea of the human playing god".
"AI suggests that machines can relieve us from mundane and hard labour but the fear is also that they can control us," Rijsdijk said.
"They are powerful and irrational."
Already we have webseries and film scripts like the completely warped sci-fi flick Sunspring, either co-written or completely written by AI. That creativity extends to music, where whole compositions have been done solely by computers.
The potential for AI is vast, and local experts talk consistently about its possible impact on health, education and other important areas of civilisation. Most believe we should develop the technology or risk being left behind.Professor Arina Britz from the Stellenbosch University node of the national Centre for Artificial Intelligence Research, explained that while AI has extreme computing capabilities, it is still limited in certain functions that only humans are capable of.CAREGIVERS
Eliminating bias is a form of reasoning that is beyond the capability of AI, according to Britz.
"Similarly, decision-making requiring autonomous desire - for example, for power - is unrealistic to expect from AI," she said.
British AI innovator Margaret Boden is concerned about using machines to perform sympathetic functions such as care-giving and working as companions.
"Domestic robots could prove to be very useful for fetching food from the fridge and completing other household tasks," she said.
"But whether an AI system can provide genuine care or companionship is another matter altogether."As with all technologies, we should be aware of the possible abuse of AI in the hands of governments and corporations, who may use it at the expense of the safety and welfare of civilians and against institutions they are hostile to.
"It is therefore not the technology we should be wary of, but people in positions of political or economic power," said Britz.
South Africa is home to the Machine Intelligence Institute of Africa and does world-leading research in the AI areas of reasoning and speech, already contributing to Google's Translate speech engine.
Institute founder Jacques Ludik said the first step to mastering AI was to understand its limitations and its strengths."The countries that don't take advantage of what you can create with AI are going to be left behind and will be dependent on other countries," said Ludik.
"We're entering into a new era, and this is something we created as humans. We are not going to stop the progress of AI or any other smart technology."If Musk and Zuckerberg are to be believed, only two possible outcomes exist for a future that fully embraces the potential of artificial intelligence.
Either an apocalyptic world run by machine-learning overlords awaits us, or we are set for a future where all of the work that does not require imagination or creativity is done by AI machines. Both of which threaten to make humans obsolete.
How technology can enhance your bodyBrain-computer interface: Merging human beings with software that allows for direct interfacing with computing devices, and improved memory.
Bionic limbs: These lifelike attachments will provide humans with increased strength, speed, balance and precision.
Ocular augmentation: Eyes fitted with lenses capable of augmented and multi-projected virtual reality environments, zoom, X-ray and infrared vision.
Sensory feedback: Tissue-synthetic interfaces connecting the biological body to a machine interacting with the cells of our bodies, exchanging information chemically, electrically, mechanically, and optically.
Powered exoskeleton: Electrically powered, wearable mobile machines that turn anyone into Iron Man.
Smart blood: Cell-sized bots in our bloodstream will monitor health, diagnose and predict illness, and detect any threats to our immune system.
Enhanced hearing: Hearing aids that record sound, generate white noise and have built-in phones.
The future is single
If you are suitably in the know you would have heard about singularity. If not, it's time you did.
In a sort of fictional sense the word has encompassed a plethora of future concepts, ranging from a kind of "when this happens, all our problems are solved" to the more reserved "technology is a tool and we should use it wisely" to the more frightful and, many feel, the more probable "Terminator 2 machine take-over apocalypse"-type scenario.
In simple terms, and using Ray Kurzweil, one of the founders of the Summit's own words, "... a future period during which the pace of technological change will be so rapid, its impact so deep, that human life will be irreversibly transformed".
The effect of exponentiality, as it were.While scientists are pushing for the UN to ban "killer robots", at the Singularity Summit held earlier this week in Joburg - and the first one in South Africa, that isn't quite the future view.
In fact, the future is rosy, abundant and idyllic. And the summit proposes that many of these developments will be in full swing in our lifetimes. Most appear to hinge on the dropping of costs, increasing access to technology and a kind of unseeable melding of tech into our lives.
These are some of the areas where you can future fit your life:
•Artificial intelligence: The Summit proposed that AI will trigger abrupt exponential technological growth - unfathomable changes to human civilisation.
• Robotics: Stronger, more versatile materials, deep machine learning, improved power systems, and associated costs to make this happen are becoming affordable, creating a space of innovation.
• Digital biology: We can now send genes by email, copying and pasting what we want. In terms of lifespans, we are getting to the point where the human lifetime is increasing by a year, every year.