The Future, This Week 01 September 2017

Introduction: 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 things that change the world. OK let's roll.

Sandra Introduction: Today in The Future, This Week: privacy (what privacy?), mind control games, and boring innovation.

Sandra: I'm Sandra Peter, the Director of Sydney Business Insights.

Kai: 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: So first story is from Gizmodo and it's titled "Facebook figured out my family secrets" and it won't tell me how.

Sandra: So this is about Facebook's uncanny ability to suggest/match friends, figure out who your contacts might be, who you might be in touch with and suggest really hundreds of people that you have come across and may have come across some of them you're not even aware that you are connected to.

Kai: So the author in the article reports that he was connected to a woman that he didn't know but on closer inspection because he was intrigued by having these people recommended to him it turned out that that was actually his great aunt who lived in a different city and they had lost contact with that arm of the family a long time ago. So he went investigating. How could it be that Facebook was able to connect him to distant family that he didn't even know about?

Sandra: And this is not the first story of its kind. Right. There are lots of examples that have appeared in the media over the past few years really weird stories such as a psychiatrist telling about how some of her patients were being recommended to each other without them actually knowing each other. Other stories of people having coffee in the same coffee shop and being recommended to each other although they hadn't met before.

Kai: So the psychiatrist story is particularly intriguing because this woman Lisa went to great lengths not to befriend her patients on Facebook. She didn't disclose her phone information to Facebook and so it was really not very clear how Facebook could make the connection among those patients which clearly would go via her. And the suspicion was that Facebook was actually using location data or it would make the connection because most of her patients had her number in their phone book. And so if those patients would disclose to Facebook their are phonebooks then Facebook algorithms would recognise the commonality that they all were connected to the same phone number and would start recommending these patients to each other which creates a massive privacy problem.

Sandra: So how is Facebook doing this? The article tried to look into the black box but how is Facebook actually gathering all this data?

Kai: You're leaving traces on Facebook by communicating, by connecting with people, by clicking likes, by using Facebook apps but also by your friends using Facebook apps and allowing the apps to actually see what you are doing by using the Like button that appears on various random websites across the Internet. But there's a whole lot more that is going on behind the scenes, invisible to the user. So Facebook is known to buy data from information brokers and that is really data about income, about credit scores, about neighbourhoods, about your credit card spend. And they can then connect that to people's profiles so they can really build much bigger profiles than what they can glean from your usage on Facebook itself.

Sandra: And there are two things we should note here: first is that there are many of these companies. So in the case of Facebook they're known to be buying data from five companies but all together there are over 90 such companies that are aggregators of that they data...

Kai:...in the U.S. alone and these companies exist worldwide, in Australia.

Sandra: Many of these companies are selling the identified data. However that's not really the case with most of these data. A couple of very interesting studies one by MIT a few years ago that looked at reidentifiability of credit card metadata and these scientists from MIT looked at credit card transactions made by about a million people and this dataset only includes things like the date of the transaction, how much money they had spent, and the name of the store. And although this information is theoretically anonymised - there are no personal details, there are no names or account numbers. The unique behaviour of people and being able to connect that to other datasets makes these random pieces of information enough to identify over 90 percent of that million people could be identified by name.

Kai: Wasn't there a famous case involving Kim Kardashian?

Sandra: Yes. A reporter at Gawker was able to identify one of the Kardashians and Ashlee Simpson and a whole range of other celebrities simply by using an anonymised database of taxi rides which are made public by the New York Taxi and Limousine Commission. So the New York taxi cab makes this data available. What the reporter at Gawker had managed to do was to connect these two Twitter feeds to Facebook pictures to Instagram posts and was able to identify these celebrities by name and exactly what taxi or limo rides they took from where to where.

Kai: So what we're saying is that each data set in itself might be anonymised but by being able to actually connect data across data sets and combine data from different sources such as IP information from where you might be using your internet, Twitter information, location data in pictures you post to Instagram, your phone number in WhatsApp and all of these kind of information you can build up and re-identify otherwise anonymous data to build quite elaborate profiles of people. And let's not forget that Facebook, as much as other companies, they own various different apps and brands. Famously Facebook when they bought WhatsApp they made the promise that nothing would change, they would not actually connect the two apps. And then they broke that promise reasonably shortly after and imported the WhatsApp contact information and phone numbers into Facebook. You could opt out at the time. Many people might not have known about it but this is how Facebook came to collect a lot of phone book data which they obviously used to make connections and recommendations for friend connections.

Sandra: The big question is why are these companies using this data to reidentify and connecting these databases.

Kai: It's a simple equation. So they have quickly figured out that more connections equals more engagement on the platform, more time spent on the platform which means more eyeball time that they can use for advertising.

Sandra: And the better ability to categorise the people that you have, to segment the market and to sell more targeted advertising and to offer that service to people who are willing.

Kai: Absolutely. You achieve two things at once: you engage people on the platform more and you learn more about them. So Facebook is known to use about 50000 different categories that they offer advertisers to micro target their messages to almost individual users of very finely targeted user groups.

Sandra: And this is working out quite well for companies like Facebook and let's not forget they're not the only ones doing this right. Google is doing this all the big five are doing this.

Kai: In fact Google and Facebook now dominate the advertising market. There's recent data which says that Google and Facebook in 2015 commandeered 60 percent of all spending that flows into online advertising. Do you know what the number is in 2016?

Sandra: Double that?

Kai: Ninety nine percent.

Sandra: That's scary.

Kai: That is very scary. So Google and Facebook very much dominate the online advertising market now.

Sandra: So with those companies controlling so much of the market and having so much data on us, privacy and the whole data ownership issue is something that we need to figure out how to do well. And indeed in a recent podcast with Hugh Durrant-Whyte, the Chief Scientist of the UK, he stressed the importance of figuring out how we think about data ownership and privacy.

Kai: So in this podcast, excellent podcast by the way, Hugh talks about all kinds of challenges around the future of work, automation and technologies and privacy is really one of them. So let's hear from the man himself.

Hugh Durrant-Whyte: I think our big thing in the future is going to have to be figuring out the whole data ownership, data privacy issue because I think actually most private companies have gone way beyond what any government do. Already. (Sandra: Is it too late?) No I don't think it's too late. I think there are interesting technologies out there that have the opportunity to force change for example as we are seeing in many cases fully encrypted communications, fully encrypted engagement with the Internet, it's perfectly possible to run the Internet and do Google searches without Google ever finding anything at all about you, if you were actually concerned about it.

Sandra: And you can listen to all of Hugh's podcast on our website or on Soundcloud, Stitcher, iTunes wherever you listen to us.

Kai: He makes an important point: data ownership is really the issue here. The control that we have or have lost about our own data. And if you look at the article it actually brings to the fore a really pertinent issue which is the author tried to inquire with Facebook about his case and how he would be connected to this woman and the kind of data that Facebook would be using and he didn't get very far because Facebook then turns the table on him and uses privacy as an argument not to disclose any information. So while clearly he feels that his privacy is invaded he doesn't actually get any chance to look into his own data that is being owned by Facebook on the grounds of privacy. So that really points the finger at the issue here.

Sandra: And even if he were able to look into the data companies like Facebook use over 100 signals to make up that recommendation so are you actually needing to opt out of God knows how many services with different companies that you would actually have to provide additional private information such as your Social Security number or your driver's licence or your passport to prove that you are who you are to then get access to whatever data they have on you.

Kai: There was an interesting case of a researcher actually trying to opt out of these information brokerage services and it turns out first of all you can't with many of them or you have to disclose so much more about yourself that it really defies the purpose.

Sandra: So what can we do about this? Let's hear a bit more from Hugh.

Hugh Durrant-Whyte: (Sandra says: How do we get people to rethink ownership?) I think you already are I think it's just a slow movement. Thirty percent of people use ad blockers. That's often kept quiet because it doesn't help a lot of people but if you went from 30 percent to 70 percent, how would you do online advertising? And then what's your business model?

Kai: Yeah so that points to a really interesting solution. The basis of all the data collection is the advertising. If more people were to install ad blockers that really would pull the rug out from under this whole data game. So really what we need is more education around what can be done in protecting privacy and not being exposed when surfing online, that people should actually give a shit, and also ad blockers might be a good start...

Sandra:...to actually address some of these problems and indeed when people care even companies like Uber can be persuaded to take a step back - so Uber used to track your location five minutes after you had closed the app. And in that decision a couple of days ago Uber removed the controversial tracking your location after your ride is over and after you've closed the app. So the company was saying that it is trying to give users back control of their own data.

Kai: Now speaking of control, our second story concerns mind control. The story is from The New York Times, it's titled "A game you can control with your mind".

Sandra: And it talks about a small start-up called Neurable and this is a company that you've actually seen in action at SIGGRAPH.

Kai: Yes so Neurable has this extension to what is virtual reality goggles. They have this how do I describe this? You know what, why don't we hear from the CEO of the company who are interviewed at SIGGRAPH.

Dr Ramses Alcaide: My name is Dr. Ramses Alcaide. I am the inventor, CEO, and president of a company called Neurable and its technologies. What it is it's a brain computer interface so it's able to interpret your brain activity. And then from there allow people to create actions. And what we're showing off today is actually a hybrid brain computer interface system that not only uses your brain but can also integrate other biometric data such as head position, eye tracking and any other sensors you want as well too.

Kai at SIGGRAPH: So when I'm in this virtual world I'm wearing these Vive glasses and there's something going on on the back of my head so what's that?

Dr Ramses Alcaide: Yeah. So the thing they are wearing is actually our headset and our headset has electrodes that basically record your brain activity noninvasively. And from there we're able to collect that data analyse it and allow you to control the experience that you're going through.

Kai at SIGGRAPH: OK. Can you describe that experience for us?

Dr Ramses Alcaide: Sure. The experience is you awaken inside a jail cell as a child with telekinetic powers, very similar ones as you would think about with like Professor X or Xavier or the girl from Stranger Things. You can pick up objects simply by thinking about them or throw them simply by willing them. You're able to basically use your brain activity to solve puzzles to enter new areas, fight robotic enemies and stop lasers with your brain.

Kai: So would that have to be trained to my brain or can I do this straight away?

Dr Ramses Alcaide: Right now we do have a training step, it takes about one minute of training, in the future we're moving that step all together.

Kai: So reading brain wave with EEG is nothing new. The innovation that the company brings is an AI algorithm that is able to distinguish between signal and noise in the brain waves that the EEG is reading. So the training step is important because it allows the algorithm to pick up when the person is focussing on certain objects. And we're talking six objects that the person has to look at. And then the algoritm learns and then you can use that in the game.

Sandra: And even that if you're talking to me while you looking at one of the objects the system gets confused.

Kai: Yeah. So that's one of the things that the author in the article mentions but it is quite amazing because just with the power of thought you can move objects on a screen and it's an area that attracts a lot of interest at the moment.

Sandra: Yep and we've all heard Tesla's SpaceX Elon Musk's latest foray into mentally controlling machines with the whole neuralink.

Kai: And there's a whole bunch of other start-ups in this space and the applications are in various areas. Neurable is significant because they are venturing into the consumer market but applications are in the medical sciences in various fields and I've asked Dr. Alcaid about this.

Dr Ramses Alcaide: Our technology started in the medical field and we used to actually work with people who had severe ALS or severe cerebral palsy. We actually used to work with wheelchairs in cars and really it has a limitless application area. And since we're primarily on the software side any company who has interest in working with our technology can simply reach out to us and you know we'll make it available to them.

Kai at SIGGRAPH: So obviously we're talking applications in the real world but we can also use this to just move things on the screen. I would imagine that if I'm able bodied in most instances my hands would be quicker and less cognitively engaging to use. But obviously there are other situations.

Dr Ramses Alcaide: Yeah no you're absolutely right. And so we're not targeting this toward people who have computers. We're targeting this toward the future of virtual reality and augmented reality where you want a hands free solution. Really the way we see it is that the future of operating systems for these next few computing platforms are going to be hands free solutions like just your voice or even eye tracking. But the thing that they're missing right now that it has been preventing users from really adapting this as their main computing platform is the lack of brain computer interfaces that allow you to tell the system when you want to take an action.

Kai at SIGGRAPH: OK so you said you used to work for the medical field - you're moving beyond that? Are you still working with that client group?

Dr Ramses Alcaide: The laboratory that this technology came from at the University of Michigan is still focussed on the medical applications. We decided to pivot toward a consumer end. And the reason for that is because it's really core to our vision which is to create a world without limitations. And if we're able to bring this to a mass consumer market it has the ability to transform the lives of millions of people who have these severe disabilities but also enrich the lives of billions of people.

Sandra: So first there are really great applications for this in the medical field. There are people with medical conditions with deafness, blindness, different types of paralysis, locked-in syndrome.

Kai: So if you're suffering from locked-in syndrome and really you cannot move at all or communicate with the outside world, if you were able to for the first time communicate with the outside world by directing objects or ideas, thoughts, concepts on a screen with your thoughts, that would be amazing.

Sandra: So this is an amazing technology to really free up these people to interact with the world. But let's turn to the consumer market now and let's look a little bit at the assumptions that this technology makes.

Kai: So the main problem here is you have to concentrate and pay attention to the things you want to move. Now this is not what we do when we are going about our daily business in our everyday world. Right. It completely neglects that most of our activity is based on embodied cognition, on embodied skill. When we're driving a car, our body is driving the car we do not have to consciously think about every action that we are performing. In fact if that was the case we would be way too slow.

Sandra: You will have to look at your brakes, you have to have your gearshift, you would have to look at everything that you're touching at your wheel and so on.

Kai: If I want to turn a light switch I just do it mostly without even noticing or thinking about it. If I have to consciously think about switching the switch every time I want to do it, this wouldn't work. So it leaves out what Kahneman calls the distinction between type 1 and 2 systems.

Sandra: So what we're talking about here is Daniel Kahneman's work and the book "Thinking Fast and Slow" which really was a new way to think about what thinking is - whether a feeling is also a form of thinking. The idea that Kahneman puts forward is that we are guided by two systems - system 1 and system 2. The system 1 is the brain's really automatic fast subconscious response to the world. It's the feeling that you have.

It's the skill that we have by way of being familiar with the world around us and being able to just use stuff without having to think about it which makes us into what we are. People who can do amazing things very quickly once you have mastered a skill like driving a car.

Sandra: So on the one hand you have the system 1 which is automatic and really intuitive. On top of that we have a much slower system that's steers what we do consciously which is the system 2 thinking and to a very large extent this is a more influential, more guiding, more conscious process that comes on top of thinking.

Kai: It's the part of our mind that allows us to reflect on the world, to stand back analyse, have new ideas, be creative, go places we haven't gone before but that's the exhausting, the slower, the more engaging part of cognition. Now the problem with these solutions like Neurable is that we have to engage that part of the brain to achieve things that we would normally do with our type 1 system and this is the problem that we see. For people who cannot engage type 1 system because they are disabled, they cannot use embodied skill as able bodied people would do, this is an amazing liberation. But for the rest of us the applications will be fairly limited to a very small subset of say interesting new games in VR.

Sandra: And speaking of interesting things let's talk a bit about boring innovation.

Kai: Our last story is from Quartz Media - it's titled "Y Combinator bets start up solving boring business problems will mint it's next great fortune."

Sandra: So Y Combinator is a very famous Silicon Valley accelerator that backed companies like Airbnb or Reddit - really famous huge names now in the market and this article talks about how VC's are now looking at really boring innovation.

Kai: They are themselves pivoting in the opposite direction to Neurable, away from shiny consumer things to the more boring business problems in the B to B market in healthcare, transportation, biotech recruiting, construction, insurance, all the kind of problems that do not make for big press and shiny presentations at conferences such as SIGGRAPH but which arguably solve a lot of important problems.

Sandra: So a good example of this is additive manufacturing or 3D printing. There was an interesting article a couple of months ago in the New Equipment Digest which is one of these industry magazines really around manufacturing and industrial professionals. They were talking about the boring lessons in innovation, the fact that the world doesn't practically change until these wild ideas meet really dull practicalities. So whilst you have 3D printing, additive manufacturing, for companies like SpaceX and GE Aviation this really leaves the bulk of the manufacturing industry out of it. And to see this 3D printing revolution really take off and really happen it needs to go to the people who it wasn't created for which is the boring manufacturing community and there is a big gap between using it for fantastic things that make the news and using it to fully transform industries. And the author was talking about the fact that he was so enthralled with all the hype and all the future seeking geekiness of these inventions that really a conversation that he had with one of these producers made him think that this is not about you know moon bases and the total end of centralised manufacturing. This was really about boring things throughput and cooling channels in injection moulding and so on, so changing the small things that will actually transform industries. And he said that this was one of the most boring conversations he's ever had. But also one of the most important lessons that he's ever had.

Kai: So one of the most significant things here is that sure the gold rush is exciting and it makes for good media but the people getting rich in the process are the ones figuring out the boring problems and selling shovels to the miners. And so we've seen a similar pivot towards the end of the dot com boom in the late 1990s /early 2000s when venture capitalists realised that a lot of the real money is in the B2B, is in figuring out the infrastructure problems and not so much the shiny front and consumer things. So maybe that is an indicator that some of the hype around consumer technology is coming to an end. And indeed there's evidence now that innovation is really slowing down in the consumer end now that we have what we've previously referred to as the Frightful Five - very large companies such as Google, Facebook, Apple, Amazon, Microsoft who control much of the B2C software development and are able to buy up new ideas and start-ups being set up just for the purpose of being bought by one of those five. So really an environment that is much different to what it was 10-15 years ago and so venture capitalists all of a sudden take more interest in solving the boring problems.

Sandra: Indeed the existence of these Frightful Five or the Fabulous Five the Big Five is mirrored in places like China where you have the BAT (the Baidu Alibaba and TenCent.

Kai: But things are different in China.

Sandra: Whilst in the consumer market many of the new start-ups are being bought out by these companies that mirror the Googles and Facebooks of the West, there is a lot of innovation in advanced manufacturing and indeed places like Shenzhen in China are considered the equivalent of Silicon Valley. But their innovation is very focussed, they're on an advanced manufacturing cluster, different supply chains and so on. So the type of innovation that we get there is the boring innovation the one that people don't really talk about.

Kai: So maybe we are on the cusp of a new wave of innovation. We have previously said that the market is waiting for the next big thing after the smartphone has now matured. Maybe that next big thing is not so much in the shiny consumer space but is indeed in the B2B, the enterprise space. It's more about hardware and more about heart problems in health in biotech, nanotech that are not easily accessible to the general public so the hype might not be as evident but clearly venture capitalists taking interest in those areas is a good sign that something might be shifting.

Sandra: So rather than waiting for the next Terminator, it's indeed about robots like we've spoken before the little Harvey robot that does something that is neither difficult nor sexy it moves potted plants around not because people wouldn't do it but because they don't do it and because it's more efficient to do that. But none of this makes the news.

Kai: And to give a final plug to Hugh Durrant-Whyte and the podcast we've previously released, a lot of the automation of innovation is going on in sectors such as agriculture, in mining, ports, not so much self-driving cars but a kind of behind the scenes infrastructure that are at work in figuring out large scale problems such as the energy revolution, climate change, how we feed a growing world population.

Sandra: And that's all we have time for today.

Kai: Thanks for listening.

Sandra: Thanks for listening.

Kai: And before we go here's a quick reminder. There's one week left in the early bird deadline for DISRUPT.SYDNEY where we are imagining the future. So a lot of the topics we talked about today will come up at the conference. Find more information on www.disruptsydney.com. We'd really love to see some of our listeners at the conference which happens on the 22nd September.

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. You can subscribe to this podcast on iTunes, Soundcloud, Stitcher or wherever you get your podcast. You can follow us online on Flipboard, Twitter or sbi.sydney.edu.au. If you have any news you want us to discuss please send them to sbi@sydney.edu.au.