Gerd Leonhard & Peter Vander Auwera: Chapter 5: how to humanise the IoT (Internet of Things)

Hello, this is Gerd Leonhard, futurist and CEO of the features agency, author of Technology vs. Humanity Today. Do another podcast about my book from late 2016, technology versus humanity, tech versus human dot com. We're doing chapter five today, chapter five in the book that's about the Internet of Things and what I call the inhuman Net of things. And my sparring partner here is, as usual, Peter Van, who's a content curator for the Futures Agency. Hello, Peter.

Hello, Gert. Hello, Gert. Chapter five, indeed, in Human Things. So it's about the Internet of Things, but the Internet of Things, it's a vast subject. So what do you consider as the Internet of Things?

Well, I think this is. It's good that we define it first. So some people call this the Internet of Everything.

I think that's GE language, or the Internet of Things, or connected everything, or smart everything. As Kevin Kelly says, it's basically to connect objects and processes and devices beyond the level of mobile Internet connections and computers. So right now we're roughly at a little bit less than 4 billion people.

Connected to the Internet on devices.

And I think latest numbers are somewhere around 40 billion or so already of Internet of Things. And that includes, of course, sensor networks. And there's a whole bunch of companies who deal with that, connecting sensors and.

Devices not in the way of a.

Cellular network, but usually a microcellular network. There's various companies doing that sort of thing. And it's basically connecting things like gas pipelines, environmental sensors, surfaces like roads, smart cities, cars, wristwatches, bridges, bridges, like everything to monitor what is going on and to increase the flow of data so that we can monitor and make decisions and predict and things like that. And many times the Internet of Things has been referred to as the next big step off the Internet. Cisco says we may have as much as 600 billion connected devices in 2025.

But, you know, in any case, it's.

Getting cheaper by the minute. So you'll find sensors already for less than a cent now, Eurocent, and they used to be 50 cents or a dollar for connected devices. And now, of course, the animals are connected. The cows are connecting here in Switzerland with a radio frequency chip, and the goods are connecting in the supermarket, and logistics are connecting, and UPS is connecting everything. So that's really what's happening, is connecting absolutely everything so that we can monitor and control and also derive intelligence.

Connecting is one thing, Gert, but what are they sensing for? Is it sound? Is it heat? Is it like, for example, the cows in Switzerland? What sort of information do they capture.

Well, in that case, it's easy because the cow wears a radio frequency ID chip around the neck. And then the cow moves up to.

The milking station and the milking station says, oh, you're number 47, and I.

Know where I can, you know, which.

Way I should apply the milk sucking devices. Right. And the cow gets milked automatically.

So it senses location in this case.

Location and form and shape and, you know, all the, the particular things that are aligned with that cow having a good experience getting milked. And those cows, by the way, provide 20% more milk than the other cows because they can choose their milking at any time. But generally a lot of things is about providing mostly very simple data. You know, temperature, for example. Or if you connect parts of the airplane engine like GE does, part of their engine philosophy is to connect the engine. So when the, when the airplane pulls up, all of the data of the engine gets transferred to a device that says, how hot has it become? And do we see fractures? And we monitor all of those things and then we can predict maintenance.

For example, talking about airplanes, there was like last week or two weeks ago, an issue with cameras, mini cameras that are built in, into the seats of every passenger. I think it was Singapore Airlines and also a US Airline, which cost a whole. So this opens a whole new can of worms related to privacy, to security. And Bruce Schneier already highlighted that in your mention in your book that this whole interconnectivity of everything, I think you have another name for that, is also opening a whole kind of worm zone in terms of security, privacy, who can look at it, who has access to the data, and so on.

Yeah, I think the biggest challenge is that it's a combination of all of those things.

Security is one thing which is obvious. As we connect everything, everything becomes more vulnerable to being connected. Like if your car is connected to the Internet for tracking purposes, then in theory I can have access to the electronics in your car.

It goes both ways.

So that's a security issue. Then there's a safety issue, like I may be misusing the information, I'm misreading.

It, or having the wrong idea about.

What it actually says, or it may be delayed by some way. That's really not security. It's more like safety, safe use and having what's called anthropomorphizing, you know, thinking that this machine is like me in terms of censoring and mistaken information. And then there's ethical issues about, you know, once I have the information from you and how Many times you've crossed this intersection, then I can have, I can deduct different information and have certain assumptions about your lifestyle, things like that. So those are vast issues. I think the main challenge for the Internet of things and what we call the smart city or smart everything, as I like to say, is that first we have the companies providing this kind of technology. They're primarily concerned with technical feasibility and price.

You know, because that's been the.

Approach of all these providers like telecoms and others for a long time. So as long as it works well and the pricing is reasonable, then they have a good business. And the ramifications of what's happening is usually not of their concern.

So they're leaving the externalities outside, for example, saying, okay, what actually happens if a city gets smart? They're looking at environmental benefits, which is great. They're looking at logistics costs decreasing. That's also great in terms of vulnerability security, but also governance of this data. They're saying, well, somebody will engage here. And many of them are not even looking at security until now. So it's an incomplete ecosystem and that.

Is really what we need to fix.

Or you get players like Google, who in. I think it's in Toronto. They built this smart city, but they basically own the whole infrastructure. They almost own the data. It's like a big monopolistic player that is not even paying for it. I mean, they are getting paid tax money to do that.

Well, I think the challenge is manifold because first, if you are the company investing in this and making it work and getting intelligence and improving things, that's a laudable thing. Right. But then this may put you in a priority position to where you are making decisions of what is okay and what's not, which. Which I think eventually needs to be changed.

So.

So once. Yeah, it's like a. It's like you inventing something and then you shouldn't be automatically the entire owner.

Of that entire food system. Yeah.

And I think that needs to be more transparency, more accountability, there to be.

More rules about that.

Also, I think there needs to be licensing rules as to who can afford this kind of technology. I mean, imagine if all of the European cities get smart and we save energy and we reduce pollution, but in Africa or South America, nobody could afford it. I mean, don't we need a licensing rule then that says, well, in that case we need to make it available very cheaply and we have to help them integrate this. So those are, those are all complex issues. I think that this whole discussion about externalities that we've had many times. You know, we need to consider as to what this technology does to the social system, to the security system, to the ethical standard, to the possibilities of society. And then we have to build safety nets and safeguards and accountability around this. Like, you know, this happens to be the, the story of the new Microsoft. Satelladela is talking about this all the time, how we invent stuff. And now we become responsible for this invention and we have to collaborate to make it really positive.

Even if you are the provider of the platform where others are building additional servers on top, as provider of the platform, you still remain responsible. That's what he's saying.

That's kind of what he's saying. At the same time, of course, he does want to continue to build augmented.

Reality tools for the Army. I mean, this is a difficult situation. I understand. But at a certain point you have to take a step and you have to say, okay, we are engaged, we're going to go all in on this. Because I don't think the Internet of things will actually be beneficial or will actually materialize if we don't solve all of the surrounding issues together. I think that's been become the approach, for example, of Cisco. That's why Cisco is talking about ethics. That's why SAP is all of a sudden talking about AI ethics. Right. And these companies are realizing that if we don't solve this, we have an incomplete ecosystem, which means very expensive, very dangerous and not really ready for big time.

Yeah, it sounds like your keyword is responsibility. Responsible ecosystems. It almost makes me think of what we had at the time, like end to end supply chain control, but then for connected devices. Is that what you're suggesting?

Well, that's a very good question. I think that what I'm suggesting initially is to say, well, if we're going to have such an invention that connects everything and makes data available, then let's think about what kind of authority we have over that data and where is it coming from and who should in the end actually open it or not open it. Otherwise we end up, you know, the exact same problem that we have on the Internet with tracking. You know, and that was considered to be important because tracking was the backbone or is the backbone of advertising, so became a necessary evil, so to speak. All of a sudden we realized, oh, Facebook is tracking us across all websites that have a Facebook like button. And that's completely different. So I mean, we shouldn't do the same thing again with the Internet of Things by saying, yeah, here's a good financial model that basically abuses data and therefore it's a credible thing to use it because that's how we're going to make money.

But it's already going in that direction, I'm afraid. If you look for example at the Google Nest device, which apparently has a built in microphone that Google forgot to mention that it has a built in microphone. But I mean you have all these devices in your house just starting there.

Yeah, you talked about the Nest. The Nest.

The Nest device.

Yeah, yeah.

So there are hidden features or sensors in those devices that we don't know about. So I would feel very cheated and I mean for me it would feel like a real attack on my privacy knowing that there is a microphone in there that can listen to me all the time without they even have mentioned it. So let's not even think about NSA and FBI and other officials tracking in another sense than advertising tracking what would happen if we have 2 billion or more of those connected devices.

Yeah, I mean, I agree with you on this challenge. There's something that's going wrong here in the sense of saying, okay, we have technical feasibility, we have a good enough price and so that's it.

Right.

Let's move forward and we'll figure out all the rest of it later. We can't do that when it's about very large structural things. This is how we got into the mess on the Internet, by saying that everything is advertising supported, therefore everything is about tracking and therefore is about complete abusive behavior of most of the major parties that do this.

I think the key point you make here is that we need those different rules because we are talking about a different level of, of infrastructure, another level of intelligence that we are creating.

Yeah, I think one of the things, I mean, I keep getting this question a lot when I speak to people at events is that why is it different now? Well, it's different now because the magnitude is so different. You know, the magnitude of Spotify tracking my listening habits. Yeah, that's, I understand that can be of a concern. But you know, this is about music. But now we're talking about every move that a good makes or that happens in nature. And that could be amazing or it could be heaven or hell. Right. It's a entirely different magnitude. And so when magnitude is this big, like crispr engineering of human genomes. Right. And changing the germline, we don't just say, well let's, you know, this is, let's try this out. Right. I mean this is like the people at the CERN laboratory saying let's make A black hole. See what benefits we can derive from it. It's like, now we can't do that here. This is a much larger story. And I think people are not seeing this kind of. They don't really think about ecosystem. They think very much about the ecosystem. Their own system of saying, like, you know, okay, it's working. Go out and sell and it will not sell. That's the whole point. It will not sell, it will not work. And people are going to push back, just like they're pushing back, for example, on the health cloud, you know, having your DNA in the cloud, because you know, who in the end will actually say that, what are the rules of that DNA and my genome available in the cloud? You know, nobody is going to safeguard that enough at this point.

You sometimes have referred to all this as the new meta intelligence.

Yeah, I think that, you know, what does that mean? Well, we have to be clear. That's part of the, of the book theme as well. It's called the global brain. Right.

Really what people are doing here, what companies are doing here, is they're building a new meta intelligence, which means connecting other intelligences into a new one. So we have the intelligence of mobile devices, of computers, of people, and now we have the intelligence of products and services and platforms and processes and the environment that goes into the same brain. Actually, Google calls that the global brain. And as we're creating this new brain, it starts to look a lot like Skynet, where everything is coming together and creating vastly powerful possibilities.

And when we have a new meta.

Intelligence, then we have to say, well, how do we democratize access to that? How do we spend the benefit? How do we control what, what data goes in and out and what the X. And that is a public function. I mean, if you're going to create an intelligence, in my view, this cannot be controlled by a company.

I mean, that would be extremely dangerous because the company is going to be always tempted to say, you know what, it's like a trillion dollars more or a trillion dollars less. Guess which choice we're going to take? I mean, that's the nature of capitalism.

So if it is not the company, who is in control, who should be in control? Gerd?

Well, just like we have today, we have rules for all kinds of things.

Including germline genomics and of course, nuclear energy and so on. We need to have global agreements as to what exactly happens in that brain, like where is the data coming to, where is it going from and where.

Is it going into, and what is.

The authority and who Guarantees my identity. And this needs to be done either by a national, local, Grammy of people or of course, international convention, a moratorium that says, okay, we have the global drain, it's extremely beneficial. But here's how we're going to control what goes in and out and who's in charge. I mean, imagine if we had the same thing again that Microsoft had in the 80s, which is a total monopoly on the OS.

They can do whatever and thankfully they.

Loosened up and it changed. But we cannot assume that. I mean, look at the Facebook problem. This is the richest company on the Internet.

How are you going to tell them.

To destroy their own business model? It's just not going to happen. We're looking at all these dozens of companies in this turf. Are they going to put humanity over technology just because it's something that we should do? That's not going to happen.

There is a difference, I think, because if we talk about control of nuclear plants, for example, that's the nuclear plant business. So those people can agree about all sorts of treaties to avoid unintended, unintended consequences. But in this case, in this meta intelligent system, it's cross industry. So it's nuclear, it's genomics, it's health, it's finance.

Well, I think we would still start in industry domains, for example, saying okay, environmental networks, sensors, oil, gas, you know.

All that information that can be grouped into something where we say okay, that becomes available in such a way, for example, we can also say anything to do with the health cloud and DNA and treatment and personalized health care and physician medicine and so on. That all goes here. And then there's a way of safeguarding it that is partly supervised by a.

Government body or by a public organization.

Like having an identity provision of a sort. I mean, we have that now for banking, right? And that's working because nobody would do online banking if it wasn't for that. And so that's something we have to build and that we have to think about. And this also goes towards transparency. Like this thing that happened with Google Nest. I mean, it's unbelievable lack of transparency that they would do these kind of things. I mean, what's next? They're going to say, well, inadvertently, we saved everything you ever said in the last half year. We're very sorry.

Yeah, I don't know.

That doesn't seem like a good idea.

I'm still stuck with the who's in control. So you would attack it or start from specific industry angles like health and finance and so on.

Well, I think that the Control part is really, it goes both ways. By saying, okay, we have built this, that is theoretically applicable as follows. And then we are touching on different domains. For example, in healthcare with the Internet of things, for example, if I connect all my devices, my Fitbit and everything, who do I trust with that information? I would trust the company that would not necessarily make money with that information, but that would sell something based around it. But still, if I would trust Apple with my remote diagnostic device, say, okay, that Apple is going to launch pretty certainly in the next few years, we're going to see an iPhone like device. That's my remote healthcare device for a hundred euros or whatever and I can cough into it, give blood and so on. Would I trust just Apple, even though I trust Apple more than others, but would I trust Apple entirely just to keep all that data? No, I would expect Apple to go and say we have a trust certificate by a body that is in charge of controlling such certificates. And that is currently done on the government, United nations, global level in different ways. And that is what we would need. We would need to have a global standard on how that is being supervised and what is security, safety, ethical, practical systems.

Yes, but if you go into the direction, not saying you said that, but if you go into the direction of United nations level, then if you look back in history, you have also organizations like the ITU and the Telcos, organizations who are trying to get this sort of almost un aura around them and they try to take to take this space. Like I mean we see it already now. I think we noticed Telefonica and Deutsche Telekom. So all these infrastructure players, they suddenly have discovered the digital ethics mantra and they are developing next deal or whatever it's called these days, which is trying to take the space.

I think that's good that they're starting to think about that.

But the question in the end does come down to this is not a.

For profit function to be in charge.

Of the ecosystem and the logic of privacy at all. I mean this is a political function, it is a cultural function, is a public function. My view is that either, for example, Switzerland would be very well positioned to create a global digital identity as a.

Product, but under government supervision. And I think a lot of these things need to be talked about and regarded as a challenge for saying, okay, we are responsible ourselves, but we also have supervision. And this would be similar to what the IEEE is suggesting. And all these bodies who deal with standards, right? Really that's what we're talking about here, standards. We need to find A way that we have standards that everybody has to stick to because that's how you maintain the quality of the system. I mean, just imagine Bruce Schneer talks about this in the great article, I think two or three years ago. I think if you just check for Bruce Schneer, like Schneier. Schneier with Click here to kill everyone. Yes, that was the headline of the article. And this could perfectly well happen because. Because all of a sudden the cars are connected, the airplanes are connected, the power plants are connected and there is ways to get in there and out there and there's so many open issues. And yeah, somebody will take advantage of that. And that's only the security part, that's not even the safety part, which is accidental misuse by users, things like that. That's called safety.

Like, you know, not driving safely and stuff. That's not security.

And then there's ethical issues.

So they all.

That's something we have to put on the table to say how we're going to address this, how we're going to make sure this is safe to use. And I mean, again, the comparison with Internet advertising is very close in many ways.

You could say that what we do.

On the Internet today with tracking and.

Advertising is not safe to use.

There was another comparison that popped in into my mind and it's obviously because of my background in financial services. But the sort of organization that you seem to suggest, and I'm using the word that you use, it has to be in the public domain, it has to be non profit. So it's almost like a non profit cooperative that has international oversight by overseers. That feels a lot like the way how SWIFT currently operates for financial transactions. The only dimension that they, in my opinion, do not control or observe is the ethical part. So it's very operational driven. So if you do this, then you cannot do that. And for security concerns, you have to do this and you have to do that. And those who don't follow the rules of the game, we can kick them out, really. But the big new dimension that I think you bring into this conversation is the ethical conversation.

Yeah. And I think that I've talked about, you know, we talked about that in every podcast. Meanwhile is the Ethics Council idea, Right. The Global Digital Ethics Council, I think needs to be done on the national or even regional level and then global in is to decide first what do we want? As Tim Cook rightly said in his video at the European Commission, that is the key question. Do we want a society where all those benefits are There. But in return I am becoming dehumanized. Or do we want to create a balance? Do we want to be inefficient for the sake of being human? I mean, these are philosophical questions. So in my view, first thing we need to do is to create such a body and say, what do we want to be? And as a majority, do we want to be human or do we want to merge into being half machine, half human, you know, singularity type thing, transhumanism. What do we want, you know, in general like 98% of us and create the bottom line for this, what kind of life that we want? Do we want to use automation to create a Star Trek society where we have to work less and we make the same money? Or what do we want? That's the first question. The second question is once we have decided what we want, we have to say what does it take to get what we want? And those are practical things like supervision of data security, safety moratoriums and so on and so on. And that goes for the Internet of things and of course artificial intelligence and then also genetic engineering and geoengineering. So we have to look at all these issues and then say, okay, here are some bottom line rules on how we do this. For example, if you build such a system that monitors everything in the smart city, the city needs to come up with a policy guideline that says you can use this data as follows and not differently. Because you know, the temptation, this is the biggest problem, right? The temptation of what happens when you have all that data. I mean, talk to Google or IBM or Cisco or of course Facebook and other social media. The temptation to use that data in new ways. I think every day there'll be billion dollar opportunities crossing your desk by saying, hey, do this because you can. And nobody's going to say no to all of it. We have to make sure that they say no because it wouldn't be accurate.

It wouldn't be allowed.

If not, we end up in not magic, but manic and toxic.

Yeah, that's an excerpt in the book. We'll talk about that next week.

But to summarize this whole thing, I am convinced that the Internet of things could be, for so many things, making things smart and connecting everything. As Kevin Kelly says, first we digitize and now we cognify. Make it smart.

Amazing.

Very good. At the same time, when things become cognitive.

Then they become they have to be governed because they're no longer stupid. And stupidity has an advantage of that, is there's less risk from the result of smartness. And that is kind of a strange thing because of course we want things to be smart. So that's kind of how I look at it. We have to think further, think globally, think also holistically from. And don't take these issues like safety and security as separate issues, but put them all together. Safety, security, ethics, privacy, you know, protection, free will, those things all go into the same bag. They're not separate things.

Did I hear you saying smartness needs to be governed?

Yes. Well, let's put it in the context like this. When everything is smart, we're going to need more than just saying, oh great, it's smart. You know, we're going to need governance of the combined smartness. And like I said before, the. The meta intelligence is the result of smartness of every piece of the system. It doesn't exist by itself. So, you know, the, the connected wheelchair is not going to do a whole lot of damage when it's just that. But if I connect it to everything, like, you know, tracking at the traffic light or the road or in a supermarket, then it can. And this is the material difference we're talking about, the amplitude and the magnitude of these possibilities, which is unfolding exponentially. And then we can just say, well, you know, what's wrong with the connected wheelchair? You know, well, nothing by itself. But when you put it all in the aggregate, as Google is very good at doing, for example, is connecting all of our data, then it's extremely powerful for good and for bad things. And we need some sort of. I mean, this role of keeping us safe is not the role of the companies. Well, it's the responsibility of the companies. But the role of keeping us safe and to creating guidelines is a public role. It's not the role of. I mean, we shouldn't. This is the same discussion about Uber or other companies becoming the wardens of public transportation. I mean, that's ridiculous. Why would you do something like that? That is a public role that is allegedly not going to make money, but spend money. Why would anybody even be interested? So this is part of the discussion I think, that we need to have about Magic, Manic, Toxic. Let's wrap up.

That's next week?

Yeah, that's next week. So you can find more stuff about digital ethics. That's our top theme here for this year at DigitalEthics co. That's our newsletter every week. Just sign up. It's only £100. No, just kidding. It's free. And then we have Digital ethics dot net.

That's a blog. I'VE been using.

And if you just Google for digital ethics and ger grd to find all my stuff, the book is@tech versus techvshuman.com. that's my book, Technology versus Humanity. It's now available in 10 languages. And you can find out more about Peter at. Peter Van Right? V A N Twitter.

That's right. Yes. Just Peter Van.

Peter Van. And of course, the FeaturesAgency.com is my company where we all work together on these issues, like 40 people that deal with future issues. And my own website, Last not least, FeatureWistgert, G-E R-.com where you can find all the details about this. Thanks very much, Peter.

Yes, thank you, Gert, for having me.

And we'll talk next week.

Talk next week.

Bye.