From Sydney Business Insights, this is our 100th podcast! This week: AI matters, hail the rides, and flying snake bots. Sandra Peter (Sydney Business Insights) and Kai Riemer (Digital Disruption Research Group) meet once a week to put their own spin on news that is impacting the future of business in The Future, This Week.
The stories this week
Other stories we bring up
Future bites / short stories
Robot of the week
Our theme music was composed and played by Linsey Pollak.
Send us your news ideas to email@example.com.
Disclaimer: We would like to advise that the following program may contain real news, occasional philosophy and ideas that may offend some listeners.
Intro: This is The Future, This Week on Sydney Business Insights. I'm Sandra Peter and I'm Kai Riemer. Every week we get together and look at the news of the week. We discuss technology, the future of business, the weird and the wonderful and things that change the world. Okay let's start. Let's start.
Sandra: Today in the future this week: AI matters, hail the rides and flying snake bots. I'm Sandra Peter, Director of Sydney Business Insights.
Kai: And I'm Kai Riemer, professor at the Business School and leader of the Digital Disruption Research Group.
Sandra: So Kai what happened in the future this week?
Kai: Well now you tell me Sandra, 100, I believe, yoo hoo!
Sandra: You're right. This is our one hundredth episode of Sydney Business Insights and many of those have been The Future, This Week.
Kai: Yeah it has been quite a journey and I got to be part of it for most of it. But we should mention that there's been a few other podcasts 'In Conversation' that I haven't been part of. There have been some really interesting ones there as well.
Sandra: We've had some fantastic guests people like Hugh White who was the Chief Scientist of the U.K. who incidentally has now become the Chief Scientist of NSW who spoke about automation, robotics and artificial intelligence and their influence on the future of business and on the future of society.
Kai: Yeah so one of my favourite episodes also one of the most listened to episode in Sydney Business Insights.
Sandra: We've spoken to Rachel Botsman about trust, we've spoken to a number of CEOs like Daniel Flynn and Clary Castrissian about social entrepreneurship, about using technology to fight poverty. We've spoken about women and the future of work and demystifying Chinese investment in Australia featuring our world class research here at the Business School. We've spoken to heads of Navy about how the Navy Fleet Air Arm does innovation, we've gone on road trips to the Powerhouse Museum to talk about 3D printing through history. Who knew it goes back to the 1700s.
Kai: And more than 70 episodes of The Future, This Week. Now at the end of its third season so we're looking forward to the fourth one next semester. It's been a fun ride and we've had a number of really interesting really important really big topics that we've discussed over the one and a half years that the two of us have been doing this.
Megan Wedge: Hey!
Kai: Oh yeah and here with us of course for the entire journey, Megan Wedge who we all know you know...
Sandra: Makes us sound good...
Kai: And keeps us honest and had to suffer through many interesting recording sessions with us. And of course we want to thank everyone at the Business School who've helped us the entire Sydney Business Insights team, Jacquelyn Hole who's told many of our stories.
Kai: And the colleagues in the Digital Disruption Research Group more broadly.
Sandra: And all of our researchers, students, collaborators who have helped us tell some very interesting stories.
Kai: And of course all of you listening to us, our audience.
Sandra: 80,000 now.
Kai: Yeah 80,000 listens. And of course we know that there's a few out there who have listened to a lot of what we had to say and this is where we want to thank you. We also want to ask you if you like what we're doing there's no harm in leaving a comment liking us on the various platforms or telling your friends all about us.
Sandra: As always you know you can find us on Spotify, iTunes, SoundCloud whereever you listen to podcasts.
Kai: And while that sounds like we are finishing up we're just getting started so what we wanted to do today is we really want to look at two of the favorite topics that have been with us week in week out that have featured in many of our episodes in one way or the other and that is artificial intelligence (AI), automation, facial recognition. Everything that surrounds that and there have been a number of stories this week again it's been bubbling up more so this week than in some of the previous weeks.
Sandra: And we'll also have another look at ride-sharing now from a completely different perspective looking at its impact on the future of cities. And of course a number of short stories we'll chat about today and we'll bring back the favorite robot of the week.
Kai: Yes. With the dragon this week. So let's get started.
Sandra: Our first story comes from the Nextweb and it's titled 'Chinese AI beats 15 doctors in tumour diagnosis competition'.
Kai: Now Sandra and I have discussed that we've been critical sometimes a bit negative on the podcast but Sandra reckons this is a good news story it features cancer and doctors losing their jobs. So why is this a good news story?
Sandra: So to me this is really a good news story. It's an AI system that's managed to outdo some of China's top doctors when it comes to diagnosing brain tumours and predicting the expansion of certain cancers. And I think it's a good news story because it's really AI at it's best. Not only doing the work better but actually complementing the work that doctors do rather than replacing it.
Kai: So the article reports on a tool called BioMind which was developed by the "Artificial Intelligence Research Center for Neurological Disorders" at Beijing Tiantan hospital and reported by Xinhua. And it is basically a typical AI application, an image recognition system that was trained with a lot of previous imaging data to recognise brain tumours in images. And it does that really really well.
Sandra: So BioMind was right 87 percent of the time compared to 66 percent by the doctors and it also took half the time diagnosing the two hundred and twenty five cases that the doctors did. Whilst predicting the expansion of such tumours, again the AI was right 83 percent of the cases compared with medical professionals who were only right 63 percent of the time.
Kai: We think this is one of those typical examples where AI really shines because the kind of deep learning technologies that we're mostly talking about when we use the term AI really have their home and originated in the field of image recognition, they go back to the famous ImageNet competition where famously a team of Canadian researchers from the University of Toronto or two people in particular Alex Krizhevsky and his collaborator Illy Sutskever came up with a way to solve the ImageNet competition which had been going on for a few years by completely changing the game reaching for deep learning at the time which wasn't an obvious choice but a neural network based technology that really solved that issue and improved image recognition accuracy dramatically and led to the kind of explosion in deep learning applications that we've seen since and these researchers famously joined Google afterwards and built what is now called AlexNet which is the application that underpins many of the advances in image recognition. So this is what AI does best and I think there's tremendous potential in medical diagnosis. Of course these systems they are not doctors right. They will not replace doctors but the potential lies in augmenting doctors, in helping doctors and in combining the expertise and skill that doctors bring with the kind of AI driven accuracy that we're seeing in deep learning image recognition.
Sandra: And it's really been this very vast datasets that we have with for instance cancer imaging data that have allowed machine learning and artificial intelligence to really thrive in some of these domains such as the one we've seen today. Now there is another side to this story which is the vast amounts of data that are not as good, not as complete, harder to analyse, harder to consolidate and there was a story in The Independent last week in which the British healthcare system (the NHS) was celebrating its 70th birthday. So it was a good time to think about how technology and automation would change the face of the NHS and we saw the same conversation there where even though we have 60 million medical records that provide the really fantastic data repository the challenges of getting that data to a state where you could have the same advances with machine learning and AI like the ones we've seen in cancer diagnosis having the same advances in things like prediction of diseases or prevention of diseases will take a very long time. These benefits are really within our grasp but there still a lot of work to be done just think of patient records that your GP might have or your hospital, some of them hand written, written to different standards, different types in different mediums or being embedded across a number of applications that some of these hospitals have. The steps we need to take to be able to have the same advances in those fields are actually the hard work that is still to be done.
Kai: So this story is really one of those forward looking stories that ask what if. What if we could leverage all of this and this is where the system should be going. But the progress of actually achieving this and this is what we want to highlight will be quite slow because of the way in which data is fragmented the quality is just not there and also the data is usually not labelled in the kind of way in which imaging data is where we have records of MRI images and we know that certain ones contain tumours and others not which we can use to train an AI if we have vast amounts of patient data there's any number of decisions that have to be made about what are we actually training the algorithms for. What are the patterns that we're training for? What is it that we want this data to do? What is it that we want the deep learning algorithms to look out for, what are the patterns? How do we employ these in treatment? But also how do we get the data to the quality level that we can actually feed into these algorithms. These are big and also boring problems.
Sandra: Indeed the article mentions the real challenges we have to face up to the fact that this is still an industry that uses fax machines and sometimes sends out letters or receives consent by letters where you could arrive to a) your medical practitioner and he would not know what medincines you've been prescribed before. They still have to ask you what you're allergic to and so on.
Kai: And also the fact that certain hospitals might have all their records in digital form, that doesn't mean that all the records for a particular patient are in that form because they go see different doctors, across different parts of the health system. Lots of records are hand written they're not integrated they are in different I.T. systems.
Sandra: So many of these challenges while grand challenges they don't make the front page. They are boring issues that have to be sorted that are so important for AI to make the same inroads. And then there is also the regulatory aspect, the policy aspect when we think about data sharing and how do we handle data across multiple systems, across multiple hospitals, in different jurisdictions and so on.
Kai: So what we want to highlight is that it's one thing to imagine what AI could be doing for us but it's another thing to actually look at what the steps are that have to be done which essentially made us think about the timelines in which we talk about this topic and the conversations that we've been having and how the topic of AI has changed in the one and a half years that we've been doing the podcast.
Sandra: In the very beginning most of the stories that we read and discussed about AI were really coming out of the tech world and were really stories where there was a lot of hype around the promise of this technology and about the advances that we managed to have, about systems beating humans and being better at recognising images or playing certain games and it was all a very positive technical conversation.
Kai: And absolutely the technological progress that has been made in the field is remarkable. This has led then to the public press picking up those topics and we've seen a lot of very overblown hyperbole like the hype stories about what these technologies promise for humanity. The dystopian stories around the robots are coming for us.
Sandra: So the way the stories have entered public conversation was really to say yes this is all fantastic but probably these things are coming for our jobs. They are going to replace us or better yet if you look at Elon Musk stories they are coming to kill us and we've really seen in the public press a fear that these technologies are fundamentally changing industries or professions in ways which are not necessarily beneficial.
Kai: Well both really. Then there's this utopian promise that AI will solve all our wicked problems for us because once we have machine intelligence and higher intelligence it can solve for us climate crises and solve the cancer problem and there's all kinds of promises that people attribute to the idea that we have intelligence in machines.
Sandra: And finally over the past few months the public conversation has really come to have a more balanced view that sees articles like the one in The Independent saying yes we've made all this progress but look at how much more we could do if we manage to solve the really boring simple problems that have to do with data. Let's remember this entire machine learning, artificial intelligence, revolution this wave of AI has come on the back of a very large amount of data that we have been able to gather.
Kai: And we've seen people in the field such as Gary Marcus who we've mentioned previously he used to be the head of AI at Uber, he's now professor at New York University who came out and basically said the hype and overblown expectations in the public media are unhelpful because they promised something that this technology cannot deliver and it's a distraction from the real progress that needs to be made in order for these technologies to become useful in areas where they are promising such as image recognition, pattern recognition, the kind of things that we've just discussed and he fears that we might be headed for another so-called AI winter. Remembering that after the first wave of AI in the 80s and 90s the whole field plunge into a long dark period of silence until neural networks and high performance computing led to a Renaissance and the kind of successes that we have today.
Sandra: So whilst we don't think it's likely that we are heading into an AI winter we are however heading into the period where we will have to do the hard work to move us forward. So this is the time for the less glamorous part of machine learning and artificial intelligence that has to do both with advances in how we handle these data but also with fundamentally we thinking how we apply this technology in our everyday life, how it's shaping our society. Let's remember in the past few months we've seen an increase in public conversation around the ethics of AI, around the uses of AI, the domains in which we let this technology make decisions for us and how we make decisions for us and we've been consistently reminded how far behind we are in terms of regulation and in terms of policy with how fast the conversation has been moving. And I think particularly telling this week we've had a article in Fortune magazine where one of the artificial intelligence brains at Google - the Google cloud CEO Diane Greene - was telling a story about how she landed at San Francisco airport and was told by the authorities that she couldn't take any pictures and if she would continue they would take her camera away.
Kai: Which is really hilarious.
Sandra: Which is indeed hilarious. She thought to herself well how are they going to get all those images out of the cloud. And the story really goes to show you that technology is moving so much faster than regulation can keep pace. We've seen this with Facebook, remember the famous answer that Mark Zuckerberg had to give during the Congress testimony "well it's ads Senator, that's how we make our money" on being asked well how do you make money if your product is free. So a lot more work to be done in the very basic building blocks of artificial intelligence and machine learning.
Kai: Yes so in essence we do think it's different this time around. The first installment of AI spectacularly failed and it failed to the point that there wasn't really much real world use. This is different this time pretty clearly because we already see a lot of applications, a lot of use being made out of deep learning, machine learning neural networks. So an AI winter is not likely. That doesn't mean that we're not going to see a potentially quite significant downturn in public perception and also in investment. There was an article just this week in VentureBeat which makes the point that what we're seeing in the AI space has all the hallmarks of the dotcom bubble a decade and a half back where a lot of the media attention, the hype, has led to a lot of investment in areas where AI might not reap the kind of benefits that are being promised at least not in the short term and that is our point.
Sandra: That is indeed the point and this is the time actually for universities to get involved, for regulators to get involved, for large sectors of the industry to actually do the hard work that is needed for this to deliver on some of the promises that it's made.
Kai: And so what we think will happen here on The Future, This Week is that while we have been quite critical in recent months of the kind of hype surrounding AI and pushed back and said this is not realistic or we've got to think of that, we have to have a look at how these things actually work, what might happen is that the hype cycle actually plunges in to a period of disinterest or AI bashing where the media comes back and rubbishes AI because some of those promises do not eventuate in the short term.
So what will probably happen is that we will have to keep attention on the topic because many of the hard problems are still unsolved, inequality problems but also regulation and the thinking about where is the impact on our daily lives once these things come online in the next five to 10 years and that also requires new approaches in universities to do research into these areas in an interdisciplinary way bringing together technical expertise, social expertise, legal expertise in what is potentially a whole new way of doing research. So while the public media might lose interest because short term gains might not be forthcoming as quickly as people think, the real research will have to happen in the years to come.
Sandra: And speaking of research, interesting research out of UCLA is Institute for Transport Studies has prompted our second story for today which comes out of Fortune Magazine titled "Ride-hailing apps may benefit poor and minority communities the most".
Kai: So this is a really interesting, positive, and counter intuitive story about research that shows that unlike taxis which have been found to often discriminate against poorer neighbourhoods, people of colour, ride-hailing services such as Uber and Lyft were found to service neighbourhoods much more equally, were also found to result in much less discrimination against minorities and different ethnic groups. So a really positive story which seemed to solve at least in Los Angeles where the research was done a particularly pernicious problem in transportation where neighbourhoods in which car ownership is low, where public transport is not necessarily the best, where ride-hailing apps actually solve a transportation problem.
Sandra: So it turns out that ride-hailing services increase overall mobility and increase it in particular for lower income neighbourhoods which would have been traditionally under serviced. And ride-sharing companies like Uber and Lyft have been in The Future, This Week and In Conversation of Sydney Business Insights podcast for quite a bit over the past year and a half including our last episode with Keith Chen the former chief economist at Uber.
Kai: Not always making the most positive news, Uber has had its fair share of problems, we've discussed previously how ride-hailing apps might take away passengers from public transport, therefore reshaping cities, adding to traffic and potentially pollution where we've seen problems with surge pricing or unfair treatment of passengers. We've reported on research that looked into collusion among drivers or unionisation among drivers and the treatment of drivers which has not always been perceived to be fair.
Sandra: But overall ride-sharing apps like Uber and Lyft have highlighted just how complex some of these new business models are to understand. And if you look at the range of stories that we've covered around Uber you will see that there were stories that covered environmental effects, there were stories around gender discrimination - turns out the algorithm that powers Uber resulted in a gender pay gap even though the algorithm does not discriminate per say but because women drive slightly slower at different times than men. On average, women Uber drivers made 7 percent less than men. We've reported on stories around the gig economy and the entire debate where different researchers try to understand how this shapes the type of employment that we see. We've talked about licensing issues and this has been in the story over the past few weeks around Uber losing its license in London which has resulted in 40,000 drivers, disputed number but we won't go into that now, 40,000 drivers losing their jobs are now regaining their jobs. There were discussions around the role of the CEO at Uber and how do we think about CEOs of startups vs. more established companies and the cult of CEO in Silicon Valley, around the success of such business ventures in places like China where we've discussed the takeover by Didi of Uber and actually Didi are now coming to Australia. So we've unveiled some of the complexity around such a seemingly simple service like Uber which has resulted in pretty much disciplines from across our university looking at the various impacts and the various aspects of it.
Kai: And indeed this week alone it's been puzzling economists in another story in Citylab - research coming out of DePaul University looked in to the use of Uber and Lyft in Chicago, asking the question what drives people to use ride-sharing over public transport when indeed the costs for public transport is so much less so these economists were puzzled by the fact that while you save time in most instances taking ride-sharing, commuting to downtown Chicago from the outer suburbs it's slightly different if you're just travelling within downtown because of heavy traffic.
But by and large you save a fair bit of time by taking ride-sharing but those gains in time come at a cost so you would typically save about 15 minutes on average but that comes at a cost of about 16/17 dollars or almost 70 dollars an hour which is a fair bit of money and so the economists try to do their numbers and put a price on every aspect, convenience and whatever, and they're still puzzled by the results that people are willing to pay such a premium for the supposed convenience of taking an Uber over public transport and so here's where I have to ask you Sandra, so you're quite a proficient Uber user - why do it?
Sandra: Turns out we don't always act rationally.
Kai: At least not according to the strictly financial terms that the economists would like to see.
Sandra: No and I think our episode with Keith Chen actually illustrated some of these behaviours.
Kai: So what Sandra's telling you: listen to that episode.
Sandra: Yes definitely listen to that episode. And indeed I have first hand experience with some of the results that Keith has come across which are really quite puzzling. Why does it make me so upset when the price surges are 2.0 and I pay double the price but when it's at 2.1 I kind of feel okay with it. I think well yeah there's probably some thinking behind it and well I guess I'd just have to pay that. Here's Keith.
Keith Chen audio: One really interesting behaviour that we saw was that people responded very differently to price increases in that surge multiplier depending upon the exact number that was framed. So think about it this way. 1.0 surge, that's exactly the normal price. 1.2 you're paying 20 percent more, 1.3 you're paying 30 percent more. And just like rational economics would suggest, 1.2, 1.3, 1.4, 1.5 demand just kind of drops. This does start to break down or what you start to see is interesting psychological phenomena. So between 1.9 and 2.0 there's a very big drop off.
What's interesting is that what the evidence seemed to suggest is that at least for new Uber riders, that 2.1 seems non-arbitrary. There must be a very fancy algorithm going on here 2.1 okay well that must be the fair price. Fine I'll pay 2.1 whereas 2.0 they're just ripping me off. Rides are twice as expensive right now? I'm not doing that. What do you take me for?
Sandra: So no we don't always act rationally.
Kai: Yeah absolutely and my point is that unlike the economic decision model we don't actually make conscious decisions every time we use a mode of transport. We have our practice. We have our habit of doing it and it's just convenient to take an Uber. And so people who can afford it without actually having to go through the cognitive burden of making the decision or am I going to use public transport or Uber will just use their phone or whatever gets them from A to B in the least burdensome way. So a lot of the time I think it's just a price that we pay for the convenience of not having to think about how we go about this aspect of our daily lives given that people are busy enough in their lives as it is so the premium that we're paying is one for peace of mind because after all we don't have to actually concern ourselves with trains running late or it being a rainy day and having to pack an umbrella.
Sandra: Or people still just working on George Street even after two years.
Kai: Yeah which you know has fundamentally disrupted Sydney life. And then there's also the aspect we used to have a neighbour who would drive to work across the city on a route where he could have saved time and money just taking the train because he just didn't like to be with other people on the train in the morning. So sometimes it's just the idea that someone will drive you from A to B and you can just enjoy your peace and quiet.
Sandra: Or enjoy meeting other people. I've been taking Uber every day for a very very long time now and I've actually seen a side of Sydney that I never thought I would: people with very interesting stories, people from different cultures with different backgrounds with other jobs, some of them who run theatres, some of them who have just moved to Australia. Some of them who just do this to meet people.
Kai: Yes so this is so different to my commute because I mean obviously I live out in the country, have a fairly long commute but what I do is I put my headphones on, I zone out, I sit on the train and I happily work forgetting my surroundings and I really wouldn't want to engage in a conversation and I don't mind to be on the train with other people but I really want to use that time to quietly work or read.
Sandra: To be fair my commute is less than 15 minutes.
Kai: Well mine is considerably more than that.
Sandra: So I think it's now time for our Future Bites, for our short stories for the week.
Kai: So Sandra what's one thing you've learnt?
Sandra: I've learnt that China's one belt to one road project is really going full steam ahead with a number of stories tackling different aspects of it and two very quickly that come to mind there is the China Belt and Road investments are now being tested in Malaysia with the big discussion going on around the very large amount that China is now investing in Malaysia with a few critics pointing to the Sri Lankan precedent where a Sri Lankan port had been built with loans from the Chinese government that the Sri Lankan government wasn't able to service and hence the Chinese state owned company took control of the Sri Lankan port. So China is currently investing very large amounts of money trying to connect the world. This is also the case with the Digital Silk Road which we've mentioned before which is part of the Chinese Belt and Road initiative. This is investment in global digital infrastructure where a discussion has emerged around the shape of the future global Internet giving the very large investment in infrastructure through the Digital Silk Road now called Belt and Router, the questions are whether China will exercise tighter control over it in the same way as it's exercising on its domestic Internet through the great firewall and its cybersecurity laws. As it turns out this conversation over the Digital Silk Road is a very very complex one from where the optical cables are laid through what information flows through these cables to also what types of algorithms will affect how that information is distributed. And this is a story that I'm sure we'll keep coming back as we are debating the many economic social security implications of these massive global infrastructure projects that China has initiated.
Kai: So my short story is also coming from China and it's another favourite on The Future, This Week which concerns self-driving cars in this instance self-driving minibuses. Baidu just made its hundredth autonomous mini bus ahead of its commercial launch in China. The article is from Tech Crunch and reports on Apolong a minibus that seats up to 14 people which is based on Baidu's Apollo autonomous driving platform which we previously mentioned, which Baidu will launch in several Chinese cities including Beijing, Shenzhen, Pingtan and Wuhan which is a so-called level four self-driving vehicle which means it's not a fully self-driving car but the vehicle can take over all driving under certain conditions so it will mainly be deployed in tourist spots, airports, and other controlled or so-called geo fenced areas so it's not a car or minibus in this instance that you can let loose in normal, messy everyday traffic so will cruise around certain areas in designated lanes or drive around certain routes that are pretty much pre-programmed where it can react to pedestrians in the lane and things like that but it's not designed to go everywhere and anywhere but for Baidu's Apollo platform it marks a big step forward. So Baidu has previously said that they are interested in creating the software platform for others to use in their self-driving car projects. So it is interesting to see that they are launching their own hardware so to speak, their own mini bus which will also soon be launched in Japan curiously.
Sandra: And this story comes on the back of the already big lead that China has in the electric bus category we've discussed this previously.
Kai: Yeah we've discussed BYD the giant Chinese manufacturer which has created huge numbers of buses already in service which dwarfs anything that we are used to in the West.
Sandra: So indeed China not only taking the lead in the number of such buses that are in circulation but also in the capacity to fulfil demand for these sort of buses across the world.
Kai: And now...
Audio: Robot of the Week.
Sandra: We couldn't have done a hundredth episode without a Robot of the Week. We haven't done one in a long time but over the many many episodes that we've had we've had everything from robots punching weeds back into the ground, to Boston Dynamic hounds opening doors and scaring the bejesus out of people, to robots at shopping malls committing suicide by drowning, to Sophia the very human-like robot that has more rights in certain countries than women do.
Kai: Yeah but this one is some scary shit again. It's called Dragon. And while Boston Dynamics robot hell hounds might be scary, this thing can fly. It's a snakelike drone that can fly. Officially called "Dual-rotor embedded multilink Robot with the Ability of multi-deGree-of-freedom aerial transformatiOM" which supposedly forms the acronym DRAGON with a little bit of language massaging.
Sandra: I don't think that was a coincidence and whilst we'll include the video in the show notes and you can see exactly how this thing goes about.
Kai: The article highlights that comprised of four propulsive thrusters, this thing can change shape to fit through different gaps and structures, it can go straight, vertical, diagonal snake. It can do them all but as the article highlights thankfully this thing can only fly for about three minutes so technically you can still out run it.
Sandra: So this is of course coming out of Japan from the University of Tokyo and whilst it looks amazing now we can't wait to see its full range of interactions which should include its two hands picking up objects like a two fingered arm wrapping itself around stuff to move it or flying out or doing other exciting things yet to be revealed.
Kai: So this thing looks like the typical thing that hasn't really figured out what it's good for.
Sandra: So definitely our Robot of the Week and that's all we have time for today.
Kai: Thanks for listening.
Sandra: Thanks for listening.
Kai: And remember to tell your friends...
Sandra: About Sydney Business Insights.
Outro: This was The Future, This Week made awesome by the Sydney Business Insights Team and members of the Digital Disruption Research Group. And every week right here with us our sound editor Megan Wedge who makes us sound good and keeps us honest. Our theme music is composed and played live from a set of garden hoses by Linsey Pollak. You can subscribe to this podcast on iTunes, Stitcher, Spotify, SoundCloud or wherever you get your podcasts. You can follow us online, on Flipboard, Twitter or sbi.sydney.edu.au. If you have any news that you want to discuss please send them to us at firstname.lastname@example.org.