ai.bythebay.io: Panel: Smart Cars and Smart Cities
Recording: ai.bythebay.io: Panel: Smart Cars and Smart Cities
Music Things that we've kind of touched upon a couple of times here today is We've heard a lot about the cars themselves, right? What's in the cars, how they work The challenges they've got But obviously These cars don't live on their own They operate in the world that we all live in The cities that we live in, the roads that we have You know, outside our homes And I think this is kind of going to be The big story about cars Now we've seen cars doing, we've seen trucks driving We've seen them on our streets Now the question is, you know, what's going to happen next? How are they going to interact with our daily lives? And particularly, you know, we've also heard that They're really likely to come into the cities first, right? Where 80% of the, you know, population lives Where probably pretty much everyone here in this room today lives And those cities are really going to be the front line Of this kind of struggle between companies to compete with each other The struggle for the technologies to succeed And, you know, the kind of way I see it I mean, obviously we can have this great view of a technological future Where everything works perfectly And the cars talk to the parking spaces And no one ever gets hit And everything just works really smoothly But, you know, obviously we haven't seen that in this early days, right? We've seen a lot of struggles We've seen struggles, you know, right here in San Francisco And so the question I really kind of wanted to ask here is You know, we've got, you know, we've got smart cars You know, we hope to have smart cities But can we have both? And will they play together nicely? And so I'll introduce the panel here We've got a fantastic group of people I'll start at this end This is Karen, Karen Tay Singapore's Smart Nation Director Covering North America Smart Nation launched in 2014 The aim of creating a future of better living for all through technology Which sounds generic but laudable Karen comes from policy and strategy in Singapore So she'll have a lot of good knowledge there Then we've got Ali Kowich Engineering Director for Advanced Automotive Systems at Delphi Here in Silicon Valley Big backgrounds in automated driving, control systems, energy storage And obviously autonomous vehicles Sitting next to him, Elliot Katz, the Global Co-Chair of Connected and Self-Driving Car Practice At DLA Piper Laws and Regulations We'll be coming to you to talk about that, Elliot Then we've got Doug Davenport Founder of Prospect Silicon Valley A non-profit urban tech innovation hub Which hits an awful lot of buttons Buzz word buttons there, doesn't it Doug? Helped startups and government agencies accelerate innovation It's already produced over $140 million in investment $50 million in public financing for trials Ted Graham, Head of Open Innovation at General Motors Working with outside partners on the future of transportation Ted's claim to fame He completed 99 rides as an UberX driver To learn about disruptive innovation in transportation Has a book coming out The Uber of Everything Which is pretty ambitious The Uber of Everything And then Will Baumgartner at the end Director of America's Transport and Mobility Business at Arup Where he leads the local planning practice Advising public and private sector clients So it's a great panel And I'm just going to open it up straight away Let's, you know, what do you know, do we have to choose between smart cities and smart cars? Can we have them both? And you know, how do we get there? Karen, your place is Can I start? Okay, so, hi everyone I'm Karen from Singapore's Smart Nation office As Will said So, can we have both? Absolutely And I think that I would expand the idea beyond smart cars It's about intelligent transportation And you know, transportation is the lifeblood of a city So if you can change that You can change how you move people and goods around the city You can change the entire city and how it's set up, right? But how we think about the idea of smart cars in Singapore is You know, what are the gains? are efficiency The gains are safety The gains are affordability And you can get these best and maximized when you deploy them at scale, right? When tons of people are riding, like everyone is riding your autonomous vehicles And that's why we are concerned about that as a government That's why we're involved in the whole move And not just letting the business models evolve And one example is Okay, how do you enable the infrastructure? How can you work with the companies to enable the WiFi, DSRC for that? You also need to work on integrating them into the public transportation system You know, they're not just functioning outside Because the public transportation system serves the bulk of people You want people in shared mobility, shared autonomous vehicles How do you incentivize that? And the final thing is also looking at Yeah, so I mean, I think those And the final thing which is really important is how you get buy-in from your people How you get everybody wanting to ride a smart car And that's not an easy thing And I think some of the panelists here will be able to talk about that as well And that is very much enabled by good regulations So, hand over the time Well, maybe jump to Elliot there Talking about regulations And you know, Singapore is quite different from the US, right? It's quite different from a lot of other jurisdictions around the world Just talk a bit about some of the challenges we're facing here Well, I think it is an important point I mean, as Karen knows Singapore is very, very different from the US I think the most significant point is that Singapore is starting from a very, very different point In the US, obviously when we talk about smart cities We talk about a post-Snowden world where a lot of people in our society believe that the government is just doing this To have another way to spy on them, right? In Singapore, as Karen knows, Singaporeans are much more trusting of the government For the most part, they think that the actions that the government takes are in their best interest Which is, at least for some, diametrically opposed to the situation in the US In terms of regulations in smart cities There's not much right now in terms of rules and regulations that are specific to smart cities One thing that we've seen recently is New York City passed Well, New York City has IoT guidelines that have been signed on also by I think about 20 other cities including San Francisco They're just guidelines They look a lot like, if you're familiar with the fair information privacy practices You're basically that for smart cities, but it's a start And I think it at least puts every city on notice that we have to roll this out in a respectable way to get consumers to buy in I mean, let's start to Doug or Will here as well I mean, we've seen some smart city initiatives, right? There are these initiatives, obviously there was the DOT smart city initiative and we've seen some kind of follow-up ones You know, talk to me about how, are those actually achieving anything? Or is it just kind of, you know, great big headlights? It's gonna achieve something Let's say that, okay? I think that, you know, we're really in very early days When you see startups come up and tell you about the dragons they slayed in order to have Their truck drive in a straight line And some of the other discussions that I've heard since I got here You know you're still in, you're trying to solve the technological problem, right? I think initiatives like the smart city challenge that was, you know, this past summer And things that are happening in other parts of the world I think that they're very important small little microcosms of what we want to see here two decades from now I think the big problem isn't necessarily are those initiatives going to accomplish something It's are we actually gonna be able to take what happens there and really effectively apply it across What's really a very complex situation? I don't know how many people know this Maybe you will or maybe you won't But just the Bay Area alone We're only, what, seven million people? How many cities are actually in the Bay Area? A hundred and ten, okay? Think about that There are so many organizations that own a tiny part of the problem That what's happening in kind of the early adopter space here in Columbus, Ohio Or in Portland or wherever it's gonna be or in San Francisco To get to scale is going to be monumental It's gonna take standards, it's gonna take communication, it's gonna take what really comes down to taking organizations That don't necessarily understand who you are or what you want to do Or what it means to really take on what it is you're building And get them to start to really capture the essence here and start to move things forward in a scalable way It's not gonna happen one city at a time, we don't have that kind of time So I think that's gonna be a huge issue One thing real quick just about specific to the DOT Smart City Challenge I think the DOT was smart to issue it as a challenge grant I believe 80 cities threw their hat in the ring And even though there was only seven finalists It did get some momentum, I think, pushing forward Cities that otherwise would not have started the process At least started getting the ball in motion Absolutely It was also really successful in getting the cities to be talking with the industry And looking for ways to partner There's so much incredible potential with what we're looking at In the space, autonomous vehicles and intelligent mobility generally You know, we've got these massive intractable problems with urban transportation all over the world And we haven't had very...I'm a transportation planner And we haven't had very many new tools in the shed for a long time on that front And now we're looking ahead to all of this potential opportunity So there's a...but there's a real need to channel that Into the right direction or help guide it in the right direction To get the right outcomes overall for society Because there's tremendous opportunity There's also tremendous risk We get mayors from suburban cities are super excited Because they're like, oh, this is gonna solve my congestion problem And solve my access to jobs problem for people that are transit dependent And that could very well be true But then we've got all the city planners and the technocrats just going Oh my god, this is everything I've been working against my entire career I've spent my entire career trying to get people out of the single occupant car And now I'm gonna have zero occupant cars driving around And what's that gonna do for congestion? So there's a tremendous amount of uncertainty There's a tremendous amount of tension And those sorts of programs at least help get the dialogue started And get us some real world experience about how they might work And just another minor point And this seems like maybe a silly point But I think it has some impact So, you know, you've heard time scales of 10 to 20 years Before you see autonomy at a kind of a system-wide scale That kind of means that the people who will be running cities at that point Haven't even thought about running for office They may not have even graduated high school yet Okay, so there's gonna be a whole new wave of what public administration, what planning, what infrastructure management, what infrastructure ownership looks like Before we get to that point There's tremendous change that's gonna happen all around this That is to me really unpredictable, so I mean it seems to me we've got things happening at different timescales, right? You're talking decadal timescales for things to come out And yet we do have people deploying cars right today, right? So is there a mismatch here between, you know, valley time Which is kind of in this sped up Keystone Cops kind of view of technology And the sort of longer term view, you know, the view of the cities, you know, as they evolve, you know, slowly to deal with that And what kind of tensions do we get there? I just want to also emphasize the role and importance of having these pilot programs like we do in Singapore I think this is a good way to take measured steps towards showing a definite improvement And I think we have to be, and I think we all here are, here for the long run I think we all understand the scale of the problem And it is going to take a lot of persistence to get there Because it is such an impactful and transformative technology for cities We have to understand that it is going to take time And that shouldn't discourage us from going full steam But just that full transformation will take time And I think, you know, trials are really important in levelling up the regulators' understanding Right? You talk to companies, self-driving companies None of them will be able to tell you what they think the regulations should be So much less so people who work in the government and, you know, don't work directly on these issues That's why when we work with Delphi, we work with Neutonomy in these trial areas And we share the data, we go through, if there is an accident, we talk about what happened And we think with them very closely about what the regulation should be, right? And what we have done recently is to put in place a regulatory sandbox for self-driving car companies Which means that, you know, we can dynamically change the legislation without going through parliament And that speeds things up The Minister of Transport can do it himself It's all in the subsidiary legislation And that's really important for companies Because you don't want your legislation to be hindering their growth Or hindering their technology So that's some things to consider when we talk about smart regulation of emerging technologies Yeah, I think Karen brings up an important point While probably no one can tell you exactly what the regulation should look like Because the technology is still in its infancy We've come a very, very long way But it's still in its infancy I think across the board what they will say Is that we don't want regulations that will box us in and ultimately stifle innovation And I think the DOT in their rollout of the AV guidelines While there's some problems If you step back and take a kind of 40,000 foot look What they did I think was very, very smart And having flexible guidelines that they're going to revisit Speak to those in industry, etc. Get feedback and do another draft, essentially To your question though about when we will actually see self-driving vehicles I think that term maybe self-driving vehicles is somewhat part of the problem Because when you say to me self-driving vehicle it doesn't really mean anything I have to know if you're talking about SAE level 3, level 4, or level 5 And what I think we'll see in 2021 Or what you know a lot of traditional automakers have been saying is level 4 Which means a vehicle where yes, you can be in the car and the vehicle will drive itself But it can only do so in a certain operational design domain Which means you know in a certain geo-fenced area under certain weather conditions or in a certain lane But I do think we'll see that relatively soon here And I mean even to add to that uncertainty on the timing It's you know there's the dimension of the what SAE level when But there's also massive questions about adoption rate Public versus private operation What the policy response is in terms of managing access to the roads And all of these things are just massive uncertainties that cities are faced with thinking about right now And they're starting to wake up now and recognize hey I'm doing a 30 year transportation plan And I don't have AVs in it, I better do something about that So that's a real grappling with that uncertainty is a huge challenge So if anybody has the answer just shoot me and let me know Because we're trying to do planning in a different context now Where we really don't know where the outcome is We don't have lines to extrapolate like we have in the past Or calibrate existing models to So we have to think about how we do planning in completely different ways And think through all these different potential combinations and scenarios that might play out I would tend to agree with your timeframes I think Delphi has been publicly committed to having a level 4 production ready technology programs by the end of 2019 And when I talked about going full force That is a big, you know, that is still a big effort to get there by 2019 So we might see it in say 2020, 2021 timeframe And just to add to that, will the regulations be there so that we can actually get these cars on the road? I mean it's, we'll see Ted, tell me about the manufacturers, the OEM's point of view I mean do you see this, you know, this connection, this collaboration with cities and jurisdictions and regulators Is that like, do you see that as a threat or as an opportunity? I mean do you want to get, you know, if you can, you know, go all in with these guys Can you give yourself a commercial advantage and can you offer a better product? Or is it a matter of just playing a defensive game? Well, I come at it from having experience outside of the automotive industry Before I started, I worked for PricewaterhouseCoopers as a consultant And I ran something called the decongestant hackathon in Toronto Which was meant to tackle congestion in a very large city in Canada And, you know, we looked at it from the point of view of bringing together policy makers, designers, people in behavioral science as well as the technologists And it really wasn't, you know, it was early signs that I was interested in how transportation was going to change But I think for the OEM's, we're not going to be just talking about AV's I think there's going to be multi-modal discussions We've got very good market research on the penetration of electric bicycles in Asia and Europe And it's no secret that we're working on that There's going to be even some interesting things around how people move around cities One of the ideas that had the most traction out of the Toronto decongestant was actually understanding how people move as pedestrians on city streets And they had this thing called hack the path, which is our, you know, cold days People, if they could avoid being on city streets of, you know, preventing people from turning right onto a road Which was backs up traffic Could they be encouraged to walk by a Starbucks and sort of internal path system Where they would earn points and things like that It wasn't anything about sort of high tech It was actually just using some interesting behavioral science and a different point of view And that was actually the one that the mayor's office was most interested in So, you know, I'm going to go back to the folks at the City of Toronto later this month and talking to them about their plans And it's going to be a wide-ranging discussion AV's will obviously be part of that But I think it's going to be trying to get a sense of what is germane to that city What's the nature of their waterfront? What are they thinking of in terms of the diversity of people there? So there is going to be some sort of customization by city, I think And we need to think about that And I think that ties into your original question as well You know, smart cities versus smart cars I wouldn't look at it necessarily as versus at all You know, everyone talks about V to V and V to I But it's just going to be V to X It's V to everything is going to be connected And I think not only do OEMs have to think about it that way But everyone needs to start thinking about it that way Well, that's interesting because, you know, we haven't heard You know, V to V is kind of a poor relation, right? Or even V to X, right? The idea of a top-down investment from...in technology You know, it's taken a long time to get going And it has not really had much traction Just kind of give me an idea if you think that this is something that will continue to get funded and get support Or is this...or are we going to rely on, you know, businesses and the markets to come up with some solutions that kind of interoperate? That looks like he's ready to talk I did? Oh, sorry No, no, I did So I could tell you this, okay Just California I'll just give you that as an example California decided certain things were going to happen In order to start to drive the carbon footprint of California down, okay We won't worry about whether you believe in climate change and all of that other stuff But they said, we're going to do this stuff And one of the things they said is they want to see higher and higher rates of electric vehicle drivers But in order to get that they knew they needed infrastructure So the state of California put several hundred million dollars into building publicly available charging infrastructure across big cities, small cities, lots of different places They did the same thing for hydrogen fuel cell vehicles, right? In this case subsidizing building of hydrogen charging stations or hydrogen fueling stations Some really interesting things and this is going on today I have every reason to believe that the state of California will see that, you know, intelligent traffic systems And kind of the enabling technology behind automated vehicles is a fundamental problem that the state has to actually take leadership on And that they'll actually look at investing similarly in those areas I think they're going to be kind of a first mover in that in some respect But that doesn't mean that the story is done I think it's wise for the people, you all developing this technology to be Not depending on a large scale V to I think standardization regulation is definitely something the government can do to enable that But in terms of paying for it, you know, California has a $300 billion deficit for infrastructure, for transportation infrastructure We need $300 billion to fix what we have and build what we need There's so much unmet need in just maintaining the infrastructure that we have Coming in and, you know, changing out traffic control systems citywide is a kind of generational project right now So there's got to be a business case and there may be a business case to do that But unless there's a business case, the government is not going to do it, they just don't have the money Singapore is in a little bit of a different stage of development, a different situation now So we think a lot about investing in this infrastructure And it's not just about building new ones, it's actually about making sure that the existing infrastructure feeds into a common database Interoperable database that can then be shared with developers and the public, right? And use because data is really what's going to drive the industry growth So we are actually having this ambitious plan for a smart nation platform Which is a platform of 100 million or more smart objects Everything from your sensors and your street lamps and your traffic lights to police cameras All this feeding back into a common operating system And then looking at how we can share that with the private sector And among government agencies so they don't have to collect so much, put up new infrastructure, right? So I would say it's about building new infrastructure and optimising your existing ones as well And when we talk about setting up challenges and bringing companies in We always think about, okay, what data can we provide them that they can't get anywhere else? Because that is our role as a government, as an aggregator How can we do that? One thing we would love to talk to self-driving car companies on is something that's coming up in the next five years We are introducing this thing called electronic road pricing version 2 Where every car in Singapore has a little satellite and we can dynamically text them on the roads Based on the distance travelled, the time travelled and all this data will be available, right? Real time traffic conditions, everything How can we benefit the developer community out there? How can we use this to support industry growth in Singapore while furthering self-driving vehicles And intelligent traffic management? So I would say we put a lot of effort into building this infrastructure that creates the data That would go on to get industry growth and good solutions for smart nation Yeah So a couple things, I mean first of all just in case you guys didn't catch that In Singapore every car has a tracking device Not yet Will have a tracking device on it And Singaporeans are okay with that, correct? We are definitely going to have to do, we have announced the intention But yeah, we will stage it such that people are comfortable with that and put in place the privacy So I think that goes to the point I was making before about trust in the government We were just very, very different between the two But a couple comments that with regard to infrastructure I think generally For as long as cars have been around, humans have been driving them That's going to change And the infrastructure, if you want to get all the safety benefits that come with autonomous vehicles The infrastructure must change with it I think that, full stop With regard to V2V specifically In December of 2016, NHTSA had a proposed rulemaking that would, if it goes through Would make V2V mandatory starting in let's say 2020, 2021 And with regard to V2V specifically I think you almost need the government to step in and do this Because if you really want to get all of the safety benefits Then it can't just be Ford vehicles talking to Ford vehicles And GM vehicles talking to GM vehicles It has to be every vehicle on the road talking to every other vehicle on the road There's, you know, debate right now whether it should be rolled out with DSRC Or it should be rolled out with 5G But once we iron out those details I think NHTSA is moving in the right direction here I'm just thinking about the tracking device in every car And, and, and Remember when the smart meters went in a few years ago with PG&E? I mean, just to be able to wirelessly read your meter And be able to run it backwards and do all the stuff we wanted to do And, you know, people, the conspiracy theories that went around with that are just incredible So it's, it's a big uphill climb I mean, I think there's a move towards, um One of the big topics for us right now is, is how we would pay for infrastructure And, and, you know, the gas tax has been a principal way of doing it But it hasn't moved in years, it hasn't changed It's not indexed to inflation Uh, cars are getting more efficient, so it's diminishing returns And so, states now in particular are starting to look at alternative ways of charging people to use the road And, um, you know, VMT-based tolls or vehicle miles travel based ways of doing it Are much more equitable in terms of your paying as you go Um, so there's some exploration doing, to do that But there's, it raises all sorts of these questions about, about privacy Right Because you just, you need a lot more data from somebody to be able to do that It's funny though, with the vision-based systems, you know, there's a lot of license plate databases out there You don't necessarily need a tracker, you know, stuck to your car You can do a lot of recognition of these things So, you know, I think we sometimes fool ourselves into thinking that A lot more data is being collected on us than the government does, right? Like, all these big companies are collecting more data on you We're not being recorded right now, are we? No Yeah that's right, I mean, we're already, yeah, we have our phones in our pockets that are Yeah You know, arguably tracking us Um, and there are, yeah, there are a bunch of apps out there that people are already using Dashcam apps that do license plate collection and then rate drivers on there Yeah Right They're driving styles, right? So you can, you know, we're already, I think, the stance is that, that, that, that, that Nexar app has recovered over Has caught over half the license plates in Bay Area already Well, it's an interesting point that I think Elliot was, was talking about before Is, there's actually a lot more trust in this country for private companies to have that data than for the government to do it And so, you know, we all know, those of us in this room, we all know we have, we've given up, we've, you know, we've ticked the box and signed the thing And it, and it gets trickier though, right? When you're talking about a smart city, a lot of what we're really talking about here is P3s or public-private partnerships So where, who owns the data, I think, is not the right question, it's who has access to that data And it's, it's a real-time issue And, in addition to your point, I, that you mentioned about, I want to raise the point of legacy problems Because, it is, I think, a big challenge How do you deal with cars that are, for example, not digitally equipped in a modern city? I agree, the traffic, the way the traffic is regulated today is arcade It used to, you know, it was designed for when humans could see it But I think we can do a lot better, a lot faster, a lot more safer with digital communication However, the problem is, you already have in the United States, what, 200 million cars on the road or something of that order How do you transition? How do you execute that transition? And I think that's still, for me personally, it's still a little bit of a challenging problem Let's talk about, sort of, the social side of things as well In terms of, you know, do you see people raising up in arms against autonomous vehicles? You see them embracing them, right? I mean, there are separate narratives here There's narratives about losing jobs There's narratives about communities losing revenue from parking tickets, you know And, you know, other kinds of, you know, changes in the way that we go about our lives Do you think people will accept those changes that might seem really painful to them That's in their neighbours or losing their own job For the benefit of, you know, reducing accidents Which, you know, in the early years are going to be Not going to be terribly perceptible difference, right? It's going to make a very small difference until it really starts to sort of, you know, that the whole system starts to work properly I think we've generally done a poor job in communicating the coming changes to the nature of work And when Uber drivers started appearing and the conversations that cities were having with taxis Seemed to be, we're still going to be able to make everybody happy But the reality was that we probably should have had more clear conversations about what was happening to the nature of work Singapore is doing some interesting things in terms of retraining throughout life You see taxi drivers being encouraged to go into nursing Which actually increases with the rise in population And so, I think this tough conversation to have is not just sort of, oh, you know, it's going to happen to an entire country You don't have to worry about this right now There are some very real conversations you need to have about, you know, what is this going to look like? What support might the government give in terms of retraining possibilities? You know, if you think of radiology, how is radiology going to be impacted by AI? This is what used to be a very well-paying occupation And if that gets replaced by machine learning, that's a whole different investment in my education that I should have rethought some time ago So, I think we just need to be grown up with that, some of these conversations And if you look at the number of transportation jobs in all of our countries, it's going to be a huge impact I mean, I think there are two parts to the social impact One of them, you know, if you want everybody to ride your autonomous vehicle, you have to manage the psyche of your population You have to help them have a very positive view towards these vehicles, right? And in Singapore, all our trains are autonomous now, but for cars, it's different So, you know, making sure those accidents happen during testing in very limited capacities Making sure that they associate trust in the autonomous vehicle with trust in the government's ability to regulate And make sure that these are rolled out in the best way possible, that's very important The other part is making sure that they see the concrete benefits of autonomous vehicles, right? It's not just rich people driving them, not driving them, rich people owning them and being transported around in them But actually autonomous vehicles are integrated in our transportation system in a way that improves the journey for everyone It makes everyone's journey more efficient, safer, you know, cheaper even now that we take the manpower cost out of the question So we need to be able to manage the psyche and make it positive towards autonomous vehicles And also make sure they see the concrete benefits of the whole transportation system becoming more intelligent and more connected I also think that an important part of this is the narrative has to be more accurate than it is right now I think probably everyone up here on this panel was similarly to me tearing their hair out after the autopilot crash in Florida And the next day every headline on the news and every headline in even major newspapers is essentially self-driving car kills person Which couldn't be further from the truth Autopilot I would say is maybe somewhere between level 2 and level 3 but way closer to level 2 in my opinion And so you're supposed to be monitoring the driving, right? At worst And this gentleman was watching a movie and paying zero attention Treating a SAE level 2 car like it was level 4 or level 5 And the fact of the matter is if that car was you know level 4 or level 5 and had V2V that accident doesn't come within 500 yards of taking place So I think the narrative here is very very important Plus the fact that on that very same day there was 2,999 other car accidents that killed people throughout the world And we've just accepted that for whatever reason we just say okay 1.2 million people die every year We accept that that's fine even though if you do if you turn around the example It's the equivalent of 8 747s crashing every single day for a given year and killing every single person inside If we had that happen for even a week in the US we would shut down aviation as we know it But we've just said we accept this risk, that's okay, carry on I mean it certainly seems like that particular accident was basically another distracted driving accident, right? Where the guy wasn't paying attention to the road for whatever reason You know it was enabled by the technology that his car had just as most of the distracted driving accidents we have We see around us are enabled or for the technology we hold in our hand and people want to check their phones But you know that narrative comes from somewhere, right? It comes from it doesn't come out of nowhere people are sold You know these vehicles are marketed aggressively in order to and people are paying extra money for these features And so it kind of comes back to you know is it the industry's role here to kind of regulate itself and make sure it doesn't overstep the mark and overhype it Or is it the government's role to come back and say it's if you know to really control these systems and to be sure they used in a you know in a proper way It's either to mandate you know hands-on alarms instead of Tesla doing it after the fact, right? I think part of it is you know the naming conventions that the Tesla chose on on autopilot It's it's very difficult to go back and sort of say well This is the assumption I made that it was practically autopilot Now you can read that whole end-user license agreements and sort of say what did I agree to here? And what did I agree to here and what did I agree to do and not to but you know as the you know the videos were coming on YouTube of people Showing themselves in the back seats and all sorts of stunt driving and things like that It was interesting to watch you know I have to say that I think Elon Musk is it has been incredible in terms of moving this whole industry along I Myself you know try these cars and I'm very interested to see what they're doing But at the same time there's a ongoing detailed conversation about what it can do and can't do in this shoulder period Where we're expecting humans to take control is going to be a very challenging period and we're going to need to over communicate We're going to need to undersell what it does sometimes and we're going to need to talk about the consequences But I think you know that I'm not seeing distracted driving go away I'm not seeing texting and driving go away by any means So you know we were trying to do some things on that side in the shoulder period like you know We've got this teen driver program that won't let you turn on the radio unless you have your seat belt on There's things in terms of controlling the speed and things like that but I think there's going to be a lot more of the manufacturers looking at what is this balance in the next you know 5-10 years as we move to full autonomous Yeah I mean I Again just going back to the autopilot crash I mean this was an egregious use of the autopilot This is a level 2 vehicle you're not supposed to be doing anywhere near what this gentleman was doing But I think it's important to say that even level 3 where you're depending on the human to snap back out of being asleep in the backseat Or you know texting whatever he or she is doing into driving the vehicle is somewhat of a tall task I think most OEMs and new entrant tech companies are shooting for level 4 or level 5 so that we they can fully cut the human which is a very bad driver I'm sorry to say it out of the equation I would I would tend to agree with that assessment If you enter this let's say gray zone of level 3 what is the reaction time of the human that reaction time might as well be Extremely long in terms of from a technological from a technological perspective I also wanted to maybe just address your comment about societal questions and benefit versus damage that this technology could bring in I do fully understand the issue of the jobs that this would potentially threaten but you got to also the other side look at the big picture today We're on average utilizing our cars what 5% of the time so an average person has a very expensive asset that they are probably That is depreciating rapidly you know within seven or eight years that they're paying interest on and they're usually only 5% I think there's a tremendous opportunity for cost saving with shared shared ride services at that and I think people will like that convenience But the question is how to make that how to make that transition and I think that is a wider question that companies alone that are pushing this technology cannot solve on their own But this should be addressed absolutely Yeah as we know political decisions are not always rational and there's a lot of areas to potentially get caught up and I kind of feel like the safety one is going to get resolved because they're not going to be on the I think that's a lot of people in the same way and I think that's a lot of people in the same way But I think that's a lot of people in the same way that are not going to be on the same way that they're not going to be on the same way And I think that's a lot of people in the same way that are not going to be on the same way that they're not going to be on the same way And I think that's a lot of people in the same way that are going to be on the same way that they're not going to be on the same way And I think that's a lot of people in the same way that are going to be on the same way that they're going to be on the same way But it's time to start doing some numbers and actually putting some quantitative thought to what are the outcomes we want And how do we back cast from that and figure out how we get there And do you think we're well placed to do that? I mean do you think that it's a question of cities I mean it seems to be like cities need to educate themselves and industry needs to educate cities, right? Is it just a one-way street though like that? Is it just about cities skilling up? I think it's a two-way street definitely I think there's a lot of solutions out there looking for problems And I think the cities know what the problems are and are able to communicate that So I think there's something to be learned on both sides It's certainly in industry's best interest to be proactive in working with government It's in government's best interest to be proactive working with industry You know I think that's that's the sweet spot that I want to be working in is you know where can we connect? The promise of the technology you know with the real challenges that we're facing and how do we mitigate against some of the risks? You know I I think places like Singapore and even London where they've done a lot of consolidation And how they how they manage transportation are going to be a much better position to be proactive About managing the outcomes the Bay Area 120 local agencies and our 27 transportation agencies It's going to be a scramble and I you know I keeps me up at night to think it's going to take some shocks Some negative shocks to the system that are going to be painful for everybody and then and then a big you know Policy response regulatory response to address that You know it could get ugly I hope it doesn't I hope we can Find the pathway forward but it's it's just a tremendous opportunity and a tremendous risk I think city should never say oh we want this technology or that technology what cities really have to do is define an ambitious enough objective and get people to work with them So in Singapore our ambitious objective is 100% shared mobility we don't want people owning their own cars or driving their own cars on the road right? So every time you think about autonomous vehicles you know IOT anything to do with investments in the transportation space We think about how is it going to help us drive people towards shared mobility options right? And hopefully not just cars but buses mass people mover systems right? So hub and spokesmodel and things like that so it's very important for the city to set that that ambitious the ambitious objective and get people to work with them Well in in Singapore's case it's not even an ambition right? It's it's a it would be an impossibility to have all AVs just driving around the streets you don't have enough space Yeah, we just have not enough space I think that's true everywhere I think that's true everywhere and and the promise of shared mobility is is great I mean getting the fleet down the parking implications alone are just you know the most kind of obvious one in terms of really getting an opportunity to reshape our cities but all of the studies right now are projecting you know the use of of of of of cars will increase And and and the capacity questions about you know there's a lot of thought about oh it's going to improve the capacity of our roads and I and and that's to be proven out And so you know there's a there's a lot of question about how we can manage that if we increase a lot of you know we have to pull new chunks of the population access to cars at a lower cost including the time the cost of your time in the vehicle so now all of a sudden your commute can be longer because it's productive time there's a lot of pressures empty vehicle circulation there's a lot of pressures to increasing vehicle circulation which is great if you're a if you're a manufacturer but it's not so great if you're a congestion agency and you're trying to make sure the streets are moving so I would say I don't think we're interested in congestion either as a vehicle manufacturer No I did I did mean that I meant just in terms of using the vehicles of using the product more But I you know I've heard the the heaven and hell reference from from Robin Chase you know the CEO of Zipcar who sold to Avis and she talks about the autonomous future and you know whether it's single passenger or no passenger I think we're seriously pushing towards pool rides as an efficient way to go and looking for at what threshold of city size do you need to be able to do that and where do people feel comfortable being next to somebody and sharing those rides and and mostly as a hub and spoke with some of the commuter rail you know Uber did some good data showing sort of where they're connecting to major commuter rail lines and how it might replace buses that were really underutilized because they weren't going into the right neighborhoods because of our projections of where people where we thought they would live 20 years ago didn't really come true Right right right So I I think there's going to be a lot of nuances in terms of shaping that future Yeah and and and and and don't get me wrong I mean there's tremendous positive opportunity with in so many ways including I mean you were touching on it but giving people new choices you know giving people those options to take the pool if they can do it for that trip they can save some money you know new bike options active mobility options you know for me if we can find a way to get you know another person on average in every car we can nearly double the capacity of our roads right in terms of people moving capacities which we really care about so there's there's lots of upside there well let's um let's open the floor to some questions I think hopefully you guys have got some questions for the panel we see one we need to record for the mic so I'll probably grab one so please raise your hand and I'll bring it back to you thanks great panel last week the shallow report came out in New York City and really sort of you know I guess the headline is well subway ridership is down because of Uber and oh yeah and you get a billion dollars from taxicab licenses and is that at risk so this whole data discussion everybody's talked about it the need for quantitative work it's there inside Uber it's there inside Lyft it's going to be inside other sort of platforms is there any forcing function and let's not talk about Singapore it's a nice place but you know is there any forcing function that you guys see to sort of get that data out there and accessible because right now it's not I have to say that my experience running this hackathon was doing it as a third party outside of the government and outside of the players right so I did it within a consultancy and it was the very first time in Toronto that Uber had shared data with the city and we had separated out the pickup and drop off location to protect privacy but combining that we had Waze data from Google we had some of our commuter rail data and the sort of the overlay of these things had never been put together in the same way and I think what had been happening was a lot of cities were saying for you to operate as Lyft or Uber in our cities you need to give us this data or else you can't operate at all and it was kind of an adversarial data sharing as opposed to what is the problem that we're trying to solve and let's try to bring the right people together and protect privacy, protect what's proprietary but I think you do need a third party to help convene that and have a safe space to share that and find out what information is relevant Yeah, in California the state stepped in the CPC stepped in with regulation of the TNCs so the cities don't have any ability to regulate that or require data at all So we talked a little bit about vehicle to vehicle communication or vehicle infrastructure how important do you guys think that is to solve the level 4 problem versus can we go with just cameras and integrate them into the space? I'm sorry, I misheard a little bit of it how important is it to solve How important do we need vehicle to vehicle communication in order to get level 4 or is it possible to come up with technological solutions without that? Well I think that V2V as I think we alluded to earlier has been in the works for quite some time now In 1999, believe it or not, when we first started talking about this it was revolutionary technology Now I think, you know, the concern is in 2017 or really 2020 if the NHTS rule making goes through when this technology will be mandated on all new vehicles is it really a belt and suspenders type deal where the sensor technology has gotten so good that there are essentially overlapping technologies at that point but I would say right now the beauty of V2V is still for most of the technology that's on these cars for the most part, there are some exceptions but for the most part, it's still line of sight, right? and V2V would be able to prevent a crash if it's a car, two cars ahead of you, right? And so that's wildly important but will it be outdated in 2020? Will it be outdated in 2025? Only time will tell I would answer that just slightly differently and say that it depends on who you're asking okay, the reason I say that, and I'm not going to use any names but one company might say we believe that it's central to our company's future to digitize all roadways and we're going to provide the computing power that allows connected cars to operate throughout an urban core, for example another one might say I don't want that money being spent or made over there because that's not my pocket I believe that that data may have value that I've yet to really kind of realize so therefore, I would rather see this to be a fully independent system that I happen to centralize myself and I might be able to provide value that kind of replaces what the government might do Others may have other ideas, right? You know, everybody's got a different take on this and I think that it's still a very undefined thing What I can tell you and, you know, I'm somewhat putting words in the mouth of an OEM but I've heard a very famous American lobbyist group that represents a bunch of OEMs saying that, you know, many OEMs are seeing V2V in the communications that are, you know, being required in the U.S. soon to be a regulatory cost, right? So they're saying, well, wait a minute, you know this is a cost that is not matched to a revenue that sounds like profit erosion which doesn't sound like I'm going to have a job, right? And so they're looking for ways that they can extract value therefore it's in their best interest not to necessarily be all that cooperative about how that could be shared with others So there's going to be some questions In addition, so Delphi is already a provider of BSRC technology, B2 everything technology In my viewpoint, that kind of technology will increase safety by decreasing errors in interpretation decreasing the robustness of the way that cars communicate because we already do that anyway we communicate our intent through various signals but those are designed for the human and I think if you have V2V and digital communication you will increase robustness you will increase safety you will decrease latency so you could, you know, pack more cars in the road because you cannot pack more cars unless you decrease that latency in terms of car-to-car communication and car-to-infrastructure communication So I think there are definitely benefits of having that type of digital communication between vehicles and the infrastructure Does anyone else have that thing? It's getting hot up here Hi Do you all foresee cities or states or nations eventually outlawing human driving? Or you can think of it from an economic standpoint where insurance companies say humans are such a high risk compared to an autonomous car we won't even insure them in the future? I can't see it I mean, people still ride horses You know, it might be confined to certain places but that's in the really, really long term I think You know, it's hard for me when people even talk about dedicated lanes it's hard for me to envision how that would actually work I think that certainly there could be cordoned areas and geo-fenced areas and city centers and things like that where so it wouldn't be outlawed completely but it might be limited in certain zones and those zones might increase over time as the public gets more confident in the technology and the cost equation makes it to the point where it doesn't make sense for you to own your own car and drive your own car but just like riding a horse or riding a bike I think people are not always some people are always going to want to have that freedom and certainly in this country with our history and our feelings towards the car it's hard for me to envision that I don't think that any country is going to or very few countries are going to use it as a lever to implement self-driving vehicles I'm not going to say like everybody your cars are banned from next year onward so get ready for that I don't think so we are going to go the south method of showing them why actually this is better for you so in Singapore we are planning some of these towns where you can't drive in the city center you can only park your cars outside and inside it's all autonomous vehicles autonomous concepts plus public transportation, biking, walking, things like that so over time those can grow and you get that buy-in and hopefully one day people will say actually I don't want to own that car it's better for me to do shared mobility I think we've got time for one more if you want to ask the last question that would be I think the answer to the previous question is Dubai I mean the ruler of Dubai has a very advanced vision on how transportation should look like in 2020 and when you read that up you can find it on Google and please look at that I mean they go way farther than you guys have been discussing just want to leave you with that Thank you very much Thank you our awesome panelists let's give a round of applause to our panel Thank you very much