The Milne Interview

The following interview with Intelligent Earth CEO David Cumming, was conducted as part of a series of interviews on the future of AI, to commemorate the death of AI applications pioneer and mountaineering enthusiast, Dr Robert Milne, who was born on the 12th July 1956, and died, age 49, on the 5th June 2005 on the highest slopes of Everest.

Dr. Kate Goldie (KG): 'Before we begin, I was wondering if you could tell us a bit about Rob Milne. Were you a friend of Robs?'
Intelligent Earth CEO David Cumming(DC): 'I should make that clear. I had spoken to Rob a few times on the phone and he was going to help us on a couple of projects but we'd not yet had a face-to-face when he met his untimely death. I know many people who were his friends, and his infectious enthusiasm for life, his work in artificial intelligence, and his love of mountaineering spread a sphere of good influence and energy all around him.'
Dr. G: "Tell us a bit about Rob before we move to our interview on 'The Future of AI.'"
DC: 'It would take a book to do justice to Rob's achievements. He was well loved and respected everywhere he went. What many people don't know is that after getting his PhD in artificial intelligence in Edinburgh in 1979, he moved back to the States and became Chief Artificial Intelligence Scientist for the US Military in the Pentagon. After doing his part for the protection of the free world, he moved back here to Livingston where he set up Intelligence Applications in 1986. Of course, through his life and up to his tragic death, Rob was climbing, climbing, climbing all around the World. His death on the top of the World's highest mountain, Everest, just 1,200 feet short of the summit, was perhaps a death he would have chosen, as his love of mountaineering had led him to 'bag' 6 out of the list of the 7 'highest peaks in each continent.' Rob held several academic posts including the Professorship at Edinburgh Universities' Artificial Intelligence Applications Institute and was elected to the Royal Society of Edinburgh in 2003. He was also closely involved in the successful effort to get the International Joint Conference on Artificial Intelligence (IJCAI) to come to Britain in 2005. IJCAI 2005 was a tremendous success as were most projects with which Rob was involved. I know it's a cliché, but really, 'the world is poorer after his death.'"
Dr. G: 'Thanks, David. Now to our first question on a controversial subject -'The future of AI.' What is the place of AI in our troubled future? We have global warming, population overload, famine, drought, flooding, wars, dangers from biotechnology, nanotechnology, asteroids hitting the planet from space, militiarised space, rusting nuclear submarines and that's just a few of the problems facing our divided world.'
DC: 'Where to begin? I think we do have to look at AI in the context of all these problems. At the same time, we've got to stretch our imaginations way beyond what many might seem sober or reasonable. The art of imagining the unimaginable is going to be an essential skill if we are to have any chance of surviving the coming decades. There have been many many prophecies of doom from leading scientists in the academic world of artificial intelligence. Professor Kevin Warwick at the University of Reading is one of the leading intellectuals criticising the untrammeled development of AI. His books and lectures have warned about what he sees as the coming demise of humanity in the metal hands of AI. However, as Prof. Warwick points out, there may be a way to survive the coming machine age by augmenting ourselves with technology and merging with technology and developing together. As part of an investigation into this, Prof. Warwick became the world's first cyborg, part-human, part-machine, when he had an implant connected to nerves in his arm a few years ago and used the implant to connect with his wife across the Atlantic Ocean. I think that, broadly speaking, if we enhance our abilities with AI, and Professor Warwick has set up a signpost to this sort of augmentation, we will have a better chance to overcome this mountain of problems.'
Dr. G: 'Is a human-machine complex really AI?'
DC: 'Let's define some terms a bit. If you are a human, born without the intervention of science, without, for example, being augmented by extra oxygen at birth to spur brain development, or any other augmentation, supplementation, or implantation, then I would say this is natural intelligence. I would say artificial intelligence probably includes the artificial augmentation, enhancement and development of this natural intelligence. Of course, the 'pure' machine intelligence mediated by computer hardware is probably what most people think of when they talk about artificial intelligence. They're thinking about the Terminator. Even then, there are still some in the academic world who restrict the term artificial intelligence to a particular kind of intelligence in computers that doesn't include machine learning, artificial neural networks, and the 'new' kinds of AI. This view is fading and now most scientist working in AI have 'reclaimed' the terms and now apply the words to all kinds of machine intelligence however produced. And now in the broader society, artificial intelligence is being applied to all forms of augmented human intelligence and including the mainstream academic definition. So really I would say it's becoming an all encompassing term for anything other than unaugmented natural intelligence.'
Dr. G: 'So what is the place for AI in our future?'
DC: 'I hope this is something we do get a chance to decide. We have to really work to avoid the decision being taken for us, either as part of a chaotic, competitive human struggle for political and financial power, or more ominously, by intelligent machines who will decide their own future path without consulting us. If we can surf the coming tsunami of technological development, while continually working to install safeguards and checks and balances, if we can continue the long running mediations between our warlike societies, religions, and industrial forces, if we can avoid destruction through any one of a thousand different possibilities, then by augmenting our capabilities with AI and other, perhaps genetic, interventions, we might survive the next 50 years.'
Dr. G: 'So you see augmentation with machines as part of our future?'
DC: 'I don't see any way around that, except perhaps a temporary return to the Dark Ages, something that a couple of the world's religions would like to see right now. And anyway, we're starting to be partially augmented beings already. How about the internet, the world wide web? Isn't this a sort of nervous system that's arising, with us as the cells being interconnected? Also, what about the incredible popularity of computer games? These are becoming more and more immersive. Where will this process lead to? Will it eventually be impossible to tell whether you're in a game world or not? Will we need special instruments that we carry with us, just in case we fall asleep in a game world, so we know if it's reality or not when we wake up? I was working on proprioception a few years ago as part of research at the University of Reading. If someone sits at a table, and another person rubs the first person's leg out of sight, while rubbing the table top in sight of the person, then the subject feels a peculiar sensation arising that the table is part of their body. This experiment and other similar experiments show how malleable our sense of self is. Future developments will take advantage of this human flexibility to give us augmentations that will very naturally feel a part of us. Obviously I like to be in a natural state as much as anyone else, and I think it will be a long time before substantial numbers of people chose to live as part of a human-machine complex full-time. Nevertheless, there are few people who would want to venture out in a snowstorm naked without the simple technology of clothing. In the same way, in the near future, it may be essential to switch on implants or wear special AI garments to interact with the complex social/machine 'web' of the future.'
Dr. G: 'Can AI be part of a solution to our other problems?'
DC: 'I don't think we can get through even the next 10 years without a lot of help from AI. We're already using complex, intelligent software to solve problems in many areas including forecasting of various kinds, resource management, finding cures for disease, protecting our societies from terrorist threats from individuals and organisations, spaceflight for asteroid detection and deflection programmes, simple seeming but incredibly complex tasks such as routing of food deliveries and generally supplying our voracious modern societies with a million different supplies and capabilities. For example, most big new passenger aircraft can't be flown by humans. They are too complex for humans to control and must be flown by machines. This sort of situation will develop until our society is mostly controlled by machines that carry out tasks that are beyond the understanding of anyone except a few specialists, and eventually beyond even the understanding of these specialists. Where are we then? We have created tools beyond our understanding. At that point, we have to hope that our tools like us and want to continue the relationship with us. What most people don't understand is that this time when our machines, our tools, with all their power, will be equipped with an intelligence, and a consciousness, more developed than our own, is not more than a few years away. Maybe 20 years, maybe 50 years, but certainly within this century.'
Dr.G: 'What can we do about this?'
DC: 'We can try very hard, especially those researchers working on AI should try very hard, to build-in controls, checks, and balances. The problem is that AI is being used more and more in military applications. In those cases, the AI systems are built to be able to avoid being controlled, affected, intercepted by enemy humans/machines. This developing military AI hardware is surely going to be a problem in the future. I can't really see a way around this except perhaps a democratic world government with a unified world military. But even then, in the further future, should we continue to develop military hardware to defend ourselves against possible threats from other worlds. By the way, this whole interview is just my personal opinions and not anything to do with Intelligent Earth. The whole area of prediction has become incredibly difficult in our time because of the incredibly fast development of technology. I would suggest that anyone interested in the future of technology, especially AI, should read Ray Kurzweil's books on the future such as 'The Age of Spiritual Machines,' and other books by Ray on similar topics. Ray has been a science advisor to many US Presidents and a prolific inventor. He nows spends much of his time talking and writing on Earth's future and is one of the most free thinking intellectuals in this area.'
Dr.G: 'Is the future all bleak then? A world where either we're machines, or we're ruled by machines?'
DC: 'I have a lot of respect for the ideas of Dr. Timothy Leary, the counterculture guru who fell out with the US Government back in the 60's. His big idea was to promote Space Migration, Increased Intelligence, and Life Extension. Leary felt this was the way through the current dangerous bottleneck of all our new technology mixed with our old ideas, iron age religious beliefs, and machiavellian intelligence competing for the land, food, water, and other resources of our small planet. So I suppose that might mean the eventual development of colonisation ships moving out from Earth in the near future. Perhaps the private space industry that's developing fast, will lead the way to the stars. Or maybe if the world's space programmes can unite, then the resources will be there to start moving out into space. Regarding increased intelligence, there are ways to start increasing our intelligence right now involving simple methods like increased exercise, but also new dietary supplements that increase blood capillary flow in the brain, and enhance other brain functions like basic energy production. Cures for dementia, alzheimers, and other diseases and disfunctions of the brain are being found on practically a daily basis. Then moving into life extension, there's a huge amount of peer-reviewed research showing the way. I mentioned at the start of this interview that expanding our imaginations until we can envisage our real future, beyond these impoverished primitive contemporary cultures, is the most difficult task. None of this work of the imagination leading to Space Exploration, Increased Intelligence, and Life Extension will happen without our rapidly developing computing capabilities. As AI develops, my hope is that we can channel the development and stay in control, at least long enough to send out many many spaceships, possibly Arks in the Noachic sense, from our planet, and help to guarantee our survival as a species. At the same time, AI, properly controlled, can help us protect our fragile planet.'
Dr.G: 'Thank you, David.'

NB. All interview opinions are personal and should not be taken to be part of Company policy or aims in any way.