The Future of Drones and the Impact on Infosec

but this presentation right now this is Andrew Claire this is the future of drones and the impact of English thanks very much all right thanks everyone for coming I guess I'll uh I'll get started here a little bit of background uh on myself I'm actually a PhD student uh right over here at MIT I spent the past six years studying uh unmanned Vehicles both how we can actually Advance the state of autonomy of these vehicles and also the human interactions that happen when a person is actually controlling these vehicles uh and so frankly I am not knowledgeable at all about information security but what I'm here today to talk about is where we're going with unmanned vehicles in the future uh

then perhaps at the end have a little bit of a discussion about what the impact of this will be on both our information security needs as the users and developers of these systems as well as potential threats honestly that will be important for all of you to deal with in the future so a little bit of a quick overview like I said I'm going to start and kind of bear with me for a little bit because I really want to explain to you what these vehicles are capable of today uh what we've been doing in the research Community to try and Advance the capabilities of these vehicles where we're going in the future and at the end

I'll try to wrap it back into what might be important to you in terms of information security needs and applications of that so a lot of you may think of drones that term in the media as primarily for a military use but I'm here today to convince you that around the world right now there are tremendous number of uses of these vehicles uh and that's coming to America in the near future uh in the next two to three years uh the FAA is working to start to integrate on that Vehicles into the national airspace I already see a lot more of these vehicles in our building so this actually happened two weeks ago in Canada there

was a driver who actually had a rollover his vehicle was uh mentally basically had a concussion and couldn't actually find his way after he got out of the vehicle he basically got lost out in the wilderness and the Rescue Services up there were actually able to use the small unmanned vehicle to actually search for and find this person before he throws hypotherapy just one example there uh in Afghanistan right now our military is using a man Vehicles delivered cargo to actually keep our Pilots out of Harm's Way for the simple job of Simply delivering packages in Japan right now they're actually doing most of their crop dusting by unmanned helicopter crop testing is a very dirty job it's not safe for the

person that's actually on board to be actually breathing in those chemicals and so this is one thing that we could also potentially see come in the United States very soon as well tracking weather NASA is actually currently using unmanned vehicles for tracking hurricane systems actually flying into other patterns that are not safe for human parts to fly into Google actually just started funding anti-poaching efforts in Nepal and India so they're actually using small unmanned vehicles to actually find and track uh poachers and try to intervene before they actually attack some of the animals obviously we have a pirate issue going on off the Somali Coast that's where we see a lot of unmanned vehicle activity

as well the nuclear disaster in Japan we saw small unmanned Vehicles being used to actually go into a radiation area monitor the radiation see what was going on help with their recovery effort there and even entertainment we're actually going to start to see these vehicles in that domain as well so this is actually a London recently it's kind of hard to see but if any of you are trekkies any trekkies to the audience maybe thank you thank you sir oh one over there uh they actually did an advertisement for the new Star Trek movie using a fleet of small unmanned vehicles to make a Star Trek logo in the sky journalists are starting to use these

vehicles uh so this is an example of a protests in Moscow that happened recently uh and so you have some journalists here flying these vehicles up there to be able to get cheaply a view of what's going on at these protests and finally we also have scientists here in America already using these vehicles to study wildlife in uh in ways that are not disruptive uh to those animals and also to try and get a good count this is actually a very informative article so I thought I would highlight some of the things that came out of this article uh the Interior Department is has a high demand for these vehicles right now and the problem is a lack of

trained pilots essentially it's actually holding back the use of these vehicles uh he sees them as a safer alternative it actually has some stories in here about a biologist and a pilot who died doing a survey of salmon just because they're helicopter crash so why can't that be done by non-manned Vehicles they could be flown in less than ideal wedding in areas such as volcanoes where we may not actually want to venture without the vehicle and generally right now it takes two or more people per vehicle to actually quality these vehicles but we're actually doing what we're doing in our research is trying to actually lower the number of people necessarily review oh because that's

what we're currently doing let's talk about what we're doing in the research Community to advance it and I'm not trying to do this to scare you although I think this is a generally paranoid crowd when you talk about Information Security Experts uh but I also want you to know where we're going in the future so that you can actually start to think about the implications for your work so uh this is actually my lab advisor uh she is a former fighter pilot in the Navy her name is Missy coming she actually wrote a book about her experience uh and about 15 or 20 years ago she actually quit being a pilot in the Navy because she saw that her F-18

aircraft could land on an aircraft carrier better than Chica and so she sort of saw the writing on the wall and said to herself I need to go back into research and start talking about these unmanned vehicles uh so right now she's a professor here at mic and I've worked with her for the past six years researching on man vehicles and actually just this week was the very first unmanned vehicle flying off of an aircraft carrier so I think she sort of Saw 15 or 20 years ago what was coming and got got some more job security as the last speaker just spoke with so this is our lab feel free to check out our lab website

um we do a lot of work on the interactions between people and different autonomous systems um so whether it's military systems Air Traffic Control nuclear power plants very Mission critical activities where you have potentially a person supervising a very highly automated system that's what we work on so this is where we're going maybe it's a little scary to all of you but this is the vision of all of us who do autonomy research but we're going to have all these vehicles all networked to each other talking to each other uh and this is actually for military operation but I just showed you how many different types of domestic commercial applications are going to be coming as

well this is what it actually looks like right now The Operators who use these vehicles so a lot of you may be computer scientists maybe in the the user experience field that whole user experience revolution has not come to the unmanned vehicle realm we haven't figured out very well how to make systems that are usable and easy for people to actually interact with so a lot of what we do is trying to make it easier for these folks to actually control these vehicles right now they use one of these little hand controllers to actually fly them but we actually said why couldn't we actually do this with a smartphone so one of the First videos I'll show you

I'll go through a set of videos of cool unmanned vehicle things that are going on in our research I was actually developing an app so the person can actually control one of these vehicles in a much more simple manner but of course this adds automation uh and computer software is what we're doing which as we just heard was a vulnerability right and so I'm going to show you is how we're advancing autonomy but also potentially adding vulnerability to the systems my research personally is on a multi-vehicle control so if you could have actually a fleet of these vehicles doing some kind of search for example that person was lost in Canada they could have found him a lot faster if

they had multiple vehicles in the air actually coordinating the search process and actually we've demonstrated this Outdoors as well so it's not just all a nice indoor controlled environments we actually have multiple vehicles collaborating with each other for some kind of search and rescue type operation you can see here one vehicle searching for a ground Target studying imagery back to a person all highly automated and that the vehicles are actually working with each other two vehicles actually do this search so that's all good and fine and a lot of you may be saying oh okay we're not we're pretty safe right now because this is all in high altitude Outdoors GPS based environments well actually our

friends over at csail here at MIT are saying oh we don't need GPS anymore what if we actually have these vehicles flying around indoors in GPS denied environments making their own maps of the environment flying in through windows that actually do their own activities such as and so you're still probably thinking to yourself oh that's still not that bad it's just this little quad rotor I could probably like knock it out you know I could just hit it with my my own hand well now we're actually our friends at seesaw found a way to make fast fixed-wing aircraft that can do rapid mapping of their environment and actually fly around and completely unmap environments such as a parking garage at

rapid speeds so you're still probably saying to yourself oh you know like this is this isn't that bad yet I can still see this thing I know it's coming I know if it's trying to infiltrate my building or something like that it could help me find my car it would help me find your car actually if you lost it in about a half hour it could it could be uh so how about a UAV that can actually fly and perch on a building and actually stay there until it needs to be used for some use or maybe to do some surveillance uh without having to use any of the power to be flying with that entire time uh

this is a group out of Stanford who did that research uh and then Darko actually came and said no those are all too big we need to miniaturize everything that you're working on so they created a hand-sized hummingbird type uad uh that could fly like that fly in through your window come into your office maybe plug in a little bit USB key into your uh your computer uh and cause some havoc in your system all right and that was still too big so our friends up at Harvard said why can't we make one that's quarter sized uh and can actually fly and do Maneuvers not quite sure what they'll do with this yet but you can imagine putting some sensors on

there uh to both do some surveillance or uh some let's say logging into your network samples like that okay so that's my scaring you videos of what we're trying to do basically our job as researchers is to make these things reality make all the things we used to think of as science fiction start to come true in reality but where are we actually going I'm sure all of you have heard about Google cars right so I view that as a type of unmanned vehicle they're using the same kind of autonomy that we're developing for aircraft and putting it on cars so we're gonna have definitely driverless cars in the near future we're gonna have for fire uh or Urban search

and rescue applications firefighting applications we're going to start to have robotic vehicles that can do that very soon my advisor actually in the office Naval Research are building an autonomous helicopter that can do medical evacuation of soldiers in a in a battle zone or dropping off cargo completely autonomously and we're also going to have it in manufacturing right so we no longer need people driving around uh carts to move things around in a manufacturing plant and we can actually start to help people with some of the more routine activities that they do on a manufacturing line with robots such as Baxter which came from a local company founded by Rodney Brooks so in the further future I'm going to

put my take out my crystal ball for a second and start to think about where we're actually going uh the first place we're going to see these uavs are an agriculture not that big of a deal for all of us kind of out in the fields on their own uh they're actually a really funny video you should all go look up online about the burrito bomber where they actually have a vehicle that they've demonstrated can parachute a burrito down to you that you order from your smartphone may happen may not happen soon as well as a beer copter which is actually going to be used in South Africa at a festival this summer where you order beer from your

smartphone and it parachutes down to you from a local podcast as well as some low altitude photography you can imagine if you're trying to sell your home maybe it'd be really cool to have a fly around video of what your home and your property looks like they're going to be companies coming up soon that can do that for you later on I think it's going to take quite a while before the Google car and self-driving cars actually become a reality uh but I'm thinking in the next five to seven years when you may see them for taxi services and limited places at first and see where we go from there as well as automated passenger

trains 10 plus years before we ever get anything large and automated flying in our Sky because the FAA is really holding back uh regulation right now to try and keep us from going there too quickly I think that's the right thing to do for safety reasons but it's going to take quite a while before we actually miss it what about battle zones battle zones so right now I mean we are using you know uavs uh they're very not automated whatsoever they're very um remotely piloted teleoperation um the military wants to get to the point where we're using more and more automation for flying I think there's a lot of resistance I think it's good resistance to having automated decision

making on board that autonomy on board those aircraft but we're moving in that direction I mean specifically about the cargo delivery delivery um it's a terrible thing for the military oh yeah yeah um so like I said right now there there is a vehicle that's doing cargo delivery in Afghanistan that are uavs it's not necessarily highly automated but certainly is what we're doing right now and they're moving in that direction uh and the 20 plus years is actually a company local here in Boston called terrafusia I was working on personal flying cars for you so if you guys want to be like the Jetsons one day and actually have your personal flying car uh it's coming slowly but it's coming uh

as well as one day you may pay twenty dollars to fly from here to California on an unpiloted commercial airport I always like to ask people what price point would have to get down to before you would get on an unpiloted commercial here especially among my generation they're like oh anything sheep I'll do it when I asked my grandma about that she said never right so we're seeing this generational shift where eventually it's going to happen someone is going to find found the first unpiloted commercial airline so this is actually more important to you let's just regarding unmanned commercial airlines how much control uh in in just commercial airlines today how much of the piloting

is controlled by software yep so when you get on an Airbus airplane uh it is capable of taking off cruising and Landing with zero pilot input at this point and for the vast majority of the flying that we do the pilot is simply a supervisor of a highly automated system it's a matter of whether the pilot chooses or not to use that automation but it's certainly there right good question so this is probably more important to all of you which is what does this mean for uh ethical legal and information security in your field so the first question that a lot of us have is we're developing complex automation a part of my job is to actually codes for that

automation right now as a researcher frankly we don't think much about how robust or how secure or how safe or how certified this automation is in the research that we need when we actually start to put this in the field this is going to be a huge problem and I think that your field has thought a lot about how you certify complex let's say web-based Automation and that's some of the information that we're going to need some of those techniques for certification are things that we're going to need very soon when you talk about ethics responsibility and liability let's just take the Google card for a second when something goes wrong when the first accident happens with your self-driving

car who is actually responsible for the action of that self-driving car was it the operator who was literally sleeping at the wheel because that's what we thought we could do nowadays was it the computer on board yourself driving car it was responsible was it Google themselves or your company who helped build that automation or was it literally the engineer who at some point programmed that automation is now going to be legally liable something happens with that self-driving environment these are sort of the questions that we're starting to ask a lot of lawyers are getting very heavily involved right now to figure out these questions and there are also going to be questions for us in our in our aerial

unmanned vehicles and then privacy obviously privacy is a large concern when it comes to these kinds of vehicles this is a recent article that I just saw online the other day actually that a person was flying one of these little quadrators around and basically said it's my legal right to fly this around and take pictures of you and your your beautiful wife sitting outside and you can't stop and so really right now there's no legal framework in place for privacy with these vehicles and I think this is a huge thing and of course the fashion industry is getting involved and they're making anti-drone wear that you can buy for two thousand dollars right now if

you want to hide from infrared cameras feel free to go online apply to use it right now and so I'm going to put this up here because like I said I am not an information security professional whatsoever but I wanted to have an open discussion with all of you about these issues that I think are problems for us uh both in terms of the security of our vehicles right so GPS cool thing has already happened uh we believe it has already happened with an actual unmanned vehicle over Iran for to actually take one of our vehicles uh also a group out of Texas has demonstrated that they were able to actually uh essentially take over a Department of Homeland Security

uh drone that was flying doing surveillance On the Border down there they actually demonstrated that they could in fact have an air control system um when we talk about supply chain security right so these vehicles are being manufactured somewhere and securing them along that entire supply chain is a problem that we need to also figure out uh actually what the last speaker talked about Bluetooth connection is cards some GM cars actually have Bluetooth connection from the brakes literally your foot pedal to the actual brakes is a Bluetooth connection and so let's talk about that as a vulnerability for cars as well when we talk about interaction between regular cars and these unmanned vehicles and airplanes regularly padded airplanes

versus unmanned airplanes that is a big problem that we're going to have uh so then that's sort of the safety aspect of information security where we're going to need help but then there's also the sort of defensive information security so how do you as an informational security professional do counter surveillance against a small micro UAV that can actually fly in your window and do something to yourself um how do you actually protect your data when you actually have all these images that are being taken by all these different cameras and sensors and generally and I think the last speaker hit on this as well if we start to rely on these automated systems for our

farming for our um search and rescue operations for our cargo delivery if UPS and FedEx all of a sudden move to an entirely automated delivery system for delivering your packages how much are we beginning to rely on automation when we're not yet there in terms of securing our systems that's all I really have um I really wanted to open up the floor and see what you all thought uh about these issues uh whether you have someone whether this seems like an impending problem the information security Community or if it doesn't I opened the floor please I actually have another question to add to it maybe you can shed some like in the research group that is there

a precedent for this um what seems like a dramatic change in technology the coming of autumn Automation in an area that really didn't have automation before is there a precedent in history for for something similar to this it had you know a fact on us that might shed some light on what's what's coming it's a good question I mean you look at things like specifically within the aviation domain we moved from manual control links to fly by wire aircraft right so all of a sudden we we embedded a computer system in the airplane such that when I move the control column it's no longer physically connected to my elevator or my aileron that scared a lot of podcasts

for a while they literally had to build trust in that computer that it was going to do what they needed to do at high speed so that they didn't die um but that that worked people have actually got past that realized the benefits of it and they were always closed systems so I said those computers were not connected to the internet or anything like that um this is a little bit different right and this we're talking about literally networking all of these vehicles probably over some kind of Bluetooth or internet-like connection um I think it exposes us to a lot of different issues so I don't know if there's a clear pressure but um you could draw some

analogies that's a great one I'm putting a new developments so regulation like I said the only regulation that's come out in the United States at least is that about a year ago they passed an FAA reauthorization act that said that there the FAA is required to attempt to integrate unmanned vehicles in the National airspace system by 2015. the problem with that regulation is that they didn't necessarily Define what an unmanned vehicle even was they didn't Define what integration meant and so there's a lot of ambiguity right there as to what that actually means the young and Gable domain and there's a lot of open questions just something to be answered especially about privacy I think that a

lot of local communities are starting to pass some laws about unmanned vehicles and drones and what's allowed and what's not allowed but there's no there's no uniform application regulation

once this stuff becomes extremely popular do you think the people will start building empty because if you put Jam the signals right like like your example of you know somebody coming in and getting your privacy okay if you could Jam the signals like even around your house you know the other thing too he says I want to just take a shotgun and pull away the uh no I'm just saying no it's very I think people what whenever you have a technology you can have people that are who ENT technology right yeah so and the other thing too is like you say if regulations will people start passing things even you know for those reasons because all

of a sudden now you start seeing tons of these things in the air all the time and all that stuff and um you know you'd be very I'm saying you can go after the communication links that technology already exists right yeah it's probably expensive important point though illegal as well think about cell phones as annoying as they are it is punishment I think a rather dramatically punishable if you yeah you mean yeah okay yeah and they also don't want you putting them on in airplanes too right yeah I mean I think you guys have a great point I think like I said the legal system has not caught up yet to where we've taken the technology I do

think that people are certainly going to start to develop anti-defensive Technologies against us I think that the information security Community needs to think about sort of defensive measures against things as well but also on on sort of our side right so how do we the large military contractors they have very dedicated security teams who are working on securing these connections but my friend is trying to develop a startup to make these small uavs that can help farmers in Armenia his hometown you know uh his own country and he doesn't have a dedicated security team that's going to help him secure the controlling for his View and so I think that we're going to need assistance in

terms of protecting our assets and our vehicles and the people who are working around them but they're also sort of going to be a fight on both sides yeah my experience is is typically show that once the stuff starts becoming uh normally out there and we're all exposed to it at some level it's going to be about a 10 years later before any serious security is actually done correctly oh yeah um that seems a day about the going going based on that I mean like one of the things the Bluetooth connection between the brakes there's a lot of high-end expensive cars where a lot of those Technologies are in place that control large portions of

vehicles I knew a number of guys that did some research on a lot of these type of Technologies being tied into all of this stuff and they found out that uh generally all the high-end vehicles uh have a large amount of Bluetooth controls where you can gain access one example was one of their experiments they've done made an airbag yeah wow as an example of they also found out when I was I was with them in Norway they found out that the uh all the major Plus in trans systems were accessible via Bluetooth uh in the technology that the set points and controls on most of these things were editable for changing orders yeah yeah so so so the thing comes in

with you know I mean we're doing the right thing here we're starting that discussion but I don't necessarily know how many of us are going to actually be involved in making those decisions a lot of times uh that may be at the drawing board where you're at yourself uh you as the designer of the future um need to take that into consideration and that question has to be proposed onto the table so every not just the big picture you know they were unmanned Vehicles security but each individual part of the components that make up that technology this question of security has to be on the table for each one of those if not you you implement this these

technology and you know hey we can control it with this device here but uh programming it setting it up all the key components within the technology can in any individual want a piece of that be manipulated externally and what kind of controls go into preventing that I know guys that have looked these the same bunch of guys have looked into aircraft control systems and to some extent now there is some physical activity since you said it was a closed system that has to be gained uh and I know some of the research they found out that some of those Control Systems could physically be paid within the Avid people that's right not all of them but some of them that

could affect the actual airplane I don't know as we added for example Entertainment Systems uh you know we got in JetBlue and we can watch our DirecTV right now that has something in some ways the entertainment systems on the airplanes that I know of have been are separated from the primary controls the problem because I know I know of some people that have actually crashed those from the back of the seats on certain types of planes so it doesn't have any effect but it does you know those systems are separating which one of the key things are the system separate are are they prevented from interacting from one system to the other uh do the general people have access to

are we using you know the theoretical close range technology Bluetooth that I could attach to a half a mile of an antenna okay obviously your research is based on your Viewpoint and the government's Viewpoint you know are you coordinating this but are there other entities other governments that are further along than us who are actually implementing some of this um that you'll work sure I mean if I go back to some of my earlier slides so really most of these examples are not from the United States right so when you look at where most things are actually happening a lot of other countries are quite a bit ahead of us in terms of implementation when I think about the

research or the the advanced resources going on I say that the vast majority of the fastest research is still occurring here in the United States but in terms of getting experience with actually implementing these in the field you know this is India Japan London Moscow these are other places around the world of Canadian Forces that are actually using this and so my my answer is I think other countries are getting more operational experiences but I think that these Advanced research is still mostly here please I actually have some experience in this field I had a team of UOP operators that work for me in Iraq Marriott Center pretty much what I see what you're talking about here is if you

look at all of these examples I've shown you see an intensibility to be able to save them and that's what you're going to see from all these commercial entities in the United States is once they kind of come to a realization of the capabilities and then you know they buy a team of folks that kind of understands it then they don't deal with it that closely but then they realize the cost savings usually pretty soon especially when you're talking about uh you know a couple presentations we only talk about what's the value of human life you know and in this case you talk about that right here Fukushima um search and rescue operation disaster relief operations all those things where

you're not quite ready to put humans in there but you're more than well into the Prototype technology so I think that's the uh the fundamental driver it's the cost and I think once the private corporations here in the United States start seeing more than what you're going to see this through a lot faster so I did want to add one additional thing about where should uh the security Community can involve and if you see over the last year the United States Navy has put out some pretty significant papers about combining uh electronic warfare and cyber warfare so I'm sorry but um you know uavs are being looked at a lot as a way to get into networks not

via flying in the window plugging USB but through the electromagnetics so as a security Community you're not thinking about how to protect your network um from threats that aren't coming in through your wires through an electronic Spectrum whether or not you have any sort of Bluetooth or other radio antennas that would be the next place that I would be looking at protect your network from so you need to look at that and uavs is one way that you're gonna be able to do that is to be able to park them right outside your window and then shoot whatever sort of electromagnetic energy or what Whatever frequency a spectrum that you want to be able to get

into that Network or Optical thank you I mean this is what I I kind of wanted to show where the technology is going and I completely agree with your sentiment that the economics you're going to drive it's like like crazy it's really going to accelerate very quickly and exactly you're right you don't even need to physically plug in anything at this point it can all be done because I like that that's great well they go along with this point about the economics driving the utilization of uavs that's actually a problem that the security standpoint is going to hinder us because security costs money if you start increasing the cost of these prices people aren't going to use them

as much or if they find they can use it but got a few Corners they will that's why as was um noted previously security really needs to be built in from the ground up not as mad on later not as a cost that can be cut we don't have to upgrade it it works fine the way it is and I think that as I mentioned I think the larger contractors and developers are thinking about these things we've had that experiences what what I'm most concerned about are sort of the small businesses are going to start this up very soon to do all these more commercial applications and I think that they're not making father and yeah yeah just to just to go

back to your like saying new things too and you're trying to get less operators yeah but it's been so much research so MIT and others on Robotics and surroundings and you should be able to almost get down to zero operators for certain paths right those types of things yeah is that kind of stuff being looked at because I mean obviously they can you know try to look around and surroundings and use the same robotics information that's been like you know certainly so um some of the the slides that I showed earlier I cannot get it back like a lot of a lot of these are working autonomously okay so uh this system that was flying through Windows has has

no person whatsoever actually involved me okay um and so I think that that's where the research Community is trying to go a lot of our research is on a higher level decision making and very uncertain command and control environments and so I think you're still going to see an operator involved in those kind of more strategic level decisions but something like uh crop dusting or basically just doing back and forth surveillance of a crop to see what state it's at can be completely automated and that's what the direction okay please well two questions one was raised earlier about um you know where the adoption is happening and it's interesting to me that it's not happening in the US as

much as it did elsewhere so why is that and and then tying into that that it appears that we're at a at a threshold between operator tell operation and automation or autonomous and what's going to push us you know I have a feeling that's an important and obviously it's a very important threshold so you know a crop independent not involved with a big city and lots of human life and Etc I could see that being adopted obviously sooner the thing that's holding back from wider adoption is safety and concerns about you know the outcome of a mistake and you said there have been some High um notableness or whatever so uh so what's holding back the US is it

law lawsuits yeah so I mean I'll go back to my my crystal ball slide for a second so um one of the main reasons why the United States hasn't seen commercial applications is our regulatory system so this bill was just passed about a year ago authorizing the FAA to start to look in the integration of it but frankly the national airspace United States is the most complex system in the world we have the busiest Aerospace in the world right now with just all of our regular commercial and private general aviation aircraft so that's one of the reasons why we're taking our time when you look at places like Australia who has the Outback after this program the

New Testament they've done a lot more testing by working while we're doing when you look at our military how water we do have the most advanced and the best on many vehicle technology because they're able to use controlled airspace to actually do their test lab so the differences between sort of the commercial private applications versus the military applications in terms of comparing us to the rest of the world in terms of your question about automation I agree with you that we're trying and basically our lab is trying to figure out what are the appropriate places to have some kind of what we call human supervisory control where you're not necessarily directly stick and Rudder flying the aircraft but there's still a

human in the loop involved in the Control process versus what are the applications where it really truly could be completely we are looking at those questions and I completely agree with you that figuring out that that allocation between those two is a very important question please sir um I know that the national highway traffic safety administration came out this week I think and told Congress that if it wants money to be doing more research into cyber security for um not autonomous vehicles but you know sort of the accident avoidance and the sort of electronic vehicle so you mentioned the Bluetooth tire sensors and all the different so the systems on automobiles are becoming significantly interconnected and complex that it's

unclear whether they're vulnerable to attack and if so if the attack could be one of my car's injury um is there I mean my sense is that it it involves sort of dropping between the outfielders at the at the federal government level that it's not really clear who has oversight of these types of issues and if there are regulations to be made you know who should be making them I mean is there a role for more government sort of oversight of how these features are implemented because my sense now is that a lot of these features are I mean a lot of the technology is feature driven but not necessarily with a lot of thought as to

possible attack vectors or ways that these features might go wrong you have a lot of good points so my understanding of the regulatory framework for uh for automobiles is that it's all state by state uh so for example Nevada California and I believe Florida are the only three states right now where you can legally test an autonomous car or a self-driving car because those legislatures have chosen to do that kind of bring some business to their state um I I think at some point the federal government will have to get involved to sort of regulate this on a national level but currently in state by state and I think that in terms of security very few people are talking about that

when it comes to self-driving cars I think it's great that and this is getting involved in that my advisor actually just is putting out a policy paper on issues with self-driving cars and one of the papers that she just decided said that for example in Austin Texas there are people using GPS Jammers just because they they want to and while your personal GPS when you're driving around in a car will still be able to know close enough where you are to work the high Precision GPS necessary for a self-driving car won't be able to work right so how do you actually deal with that problem is something that people are starting to look into but still at a

very very early stage maybe one more question uh sure the coordination among among various regulatory is a playing card as well it can learn also so they are using in trouble to same kind of equipment

and I mean he has to separately work with the FAA and you know various state and federal Automotive regulatory bodies and they're not talking with each other

thank you very much