Autonomous Vehicle Theft: The Future of Carjacking

old school auto theft vs. autonomous vehicle theft

It’s 2034. You’re in the backseat of your driverless car, sleeping, catching up on some paperwork, cramming for a test, basically handing over the control of your smart auto to its mechanical brain. Out of nowhere, your car goes left instead of right, powers through the traffic light instead of stops and you sit helplessly as your high-tech, state-of-the-art self-driving wonder is taken over from miles away. Stealing a vehicle takes on a whole new meaning when it’s done by tapping into that artificial intelligence running your autonomous car’s brain. What do you do? And this goes beyond a question of design to law enforcement. After all, in this brand new world, what is being done about autonomous vehicle theft? Will there be a difference with how it’s handled today? Should there be?

A quick history of car theft

Let’s take a look at where “boosting” a motor vehicle started so we can understand where it’s going.

1888: All in the name of marketing

first auto theft for a good cause

Bertha Benz with husband, Karl, in one of his Motorwagens, the Benz Viktoria, model 1894

The first person to actually steal a car was Bertha Benz. Wife of Karl Benz, the creator of what many call the first practical gasoline-powered automobile in 1885 and co-founder of Mercedes-Benz, Bertha took the vehicle with the best intentions — viral marketing her notoriously shy husband’s invention. However, she still stole one of his “Motorwagens” from his shop in the dead of night on 05 August 1888 with the help of her two teenage sons without Karl’s knowledge. This first case of motor vehicle theft — what is sometimes called grand theft, auto — was harmless enough and actually benefited the automotive community by creating interest in this curious contraption traveling the countryside to the extent that others wanted one for themselves. A whole industry was born and as the horseless carriage became more the norm around the world, local communities saw this actual crime grow and began to wonder how to combat it.

1896: A baron’s ride is coveted

first auto theft on record

Baron de Zuylen de Nyevelt, 1907

The true first vehicle that was really stolen was the Peugeot owned by Baron de Zuylen in 1896. His mechanic decided to take it upon himself to take it out of the shop where it was being repaired. Luckily, both the thief and the car were found nearby.

1919: The U.S. takes a stand

the emergence of auto theft

1921 National Insurance Bureau Brochure, Courtesy of

The Dyer Act, aka the National Motor Vehicle Theft Act, becomes the first federal law combatting the brand new world of stolen motorized vehicles in America. Enacted by the U.S. Congress on October 29, 1919, Dyer focuses on the interstate trafficking or transport of stolen vehicles — stealing a car and taking it across state lines. Punishment is an unspecified fine — more than likely to be based on the actual value of the car stolen — up to ten years in prison, or both.

1969: Introducing the VIN

one of the original VINs on a car

Old School ID, 1970 Chevrolet Chevelle, Courtesy of

On January 1, 1969, The Department of Transportation, formed just a few years prior in 1966, requires all new cars imported into and/or sold in the U.S. to have a vehicle identification number (VIN). At this point, it is left to each manufacturer to decide how and where they want these to appear, which can cause confusion at times.

1980: Setting some standards

how to read a VIN for autonomous vehicle theft

Image Courtesy of “How to Decode a VIN Vehicle Number” by Jason Unrau for

1980: The universal 17-character VIN system is established.

1984: Divide and conquer goes on the road

cars chopped for parts

Photo Courtesy of

The “chop shop” starts showing up. Thieves steal a car, “chop” it up, and sell the parts that don’t have the VIN on them. So, the 1984 Motor Vehicle Theft Law Enforcement Act is created, highlighting the need to trace and recover stolen motor vehicles and their parts. The Department of Transportation (DOT) reacts to the new law by mandating that the VIN now be displayed on the car’s engine, transmission, and 12 major auto body parts. This makes it more difficult to cut up the cars and get away with the theft.

1992: Grand Theft Auto gets hands-on

how car jacking can also be done for autonomous vehicle theft

Image Courtesy of “Keep Your Wheels: 11 Tips to Avoid a Carjacking” for

1992: Then came carjacking, the armed and forced taking of a car from an individual. The Anti-Car Theft Act of 1992 makes carjacking and owning, operating and maintaining a chop shop federal offenses.

1994: “Somebody, stop me!”

added protection against vehicle theft

Bumper sticker with the blue car and eyes for wheels means your car is part of the watch program

The Anti-Car Theft Act is joined by the addition of the Motor Vehicle Theft Prevention Act to the Violent Crime Control and Law Enforcement Act of 1994. This requires the Attorney General to work with the different States to develop a voluntary auto theft prevention program that has car owners signing a consent form that authorizes local law enforcement to stop the car when operated under certain conditions, such as at night, in certain urban neighborhoods, etc. Participants display a decal on their car showing they are part of the program. Carjacking is also addressed within the body of the Violent Crime Control and Law Enforcement Act itself, proclaiming that should a death result from the action, it is a federal capital offense.

1996: The age of the database

databases incorporated to fight car and autonomous vehicle theft

Milwaukee motorcycle police officer Dennie Sanchez runs a computer check on a vehicle he stopped near N. 27th and W. Wright streets. Credit: Gary Porter

Welcome the Anti-Car Theft Improvements Act, which upgrades state motor vehicle department databases that contain titling federal and local law enforcement can access to determine whether the vehicle in that person’s possession is stolen or not.

21st Century: The future of vehicle theft

Incremental advances have been consistently made in federal laws and states have their own mandates that determine whether the stealing or tampering of a car is either a misdemeanor or felony. Since the last law was put on the books in 1996, auto theft has been on the decline, BUT…

With the advent of the Internet, the new crime of cyber-hacking has given rise to an even broader, more detrimental concern. What would happen in an autonomous vehicle theft? Carjacking in absentia means there has been no armed taking of the automobile, so it wouldn’t fall under the Anti-Car Theft Act of 1992 — that outlines carjacking as taking a vehicle from an individual by force while armed.

In addition, the ability to virtually break into a car has led to terrorist concerns. Throughout the world, automobiles are used for suicide attacks and the ability to do so from afar has prompted some serious research on how to get in front of what has yet to but could happen. What, then, would that protection look like and how would law enforcement engage with, let alone track down, who took over your vehicle?

Protection against cyber-jacking

cyber attack for autonomous vehicle theft

Vehicles used as weapons of terror appear to be on the rise. From disenfranchised individuals driving down crowded sidewalks to radicals using them as bombs, the humble motorcar has taken on an ominous feel of late. It’s not surprising that self-driving and highly connected autos make many wonder just how vulnerable their inner workings will be to tech-savvy terrorists or simple car thieves.

While there have been instances, in testing, of white hat hackers infiltrating cars in order to demonstrate how these automobiles might be taken over, the actual use of a vehicle in a terrorist attack by virtue of cracking its computer code has yet to occur. However, preempting that is key and such minds as American University Computer Science Professor Nathalie Japkowicz are working with others to do just that.

The best defense is a good (computerized) offense

While there is legislation currently considered to protect us in the not-so-distant fully autonomous future, anti-theft technology is being pursued right now. Professor Japkowicz — along with Adrian Taylor of Defence R&D Canada and Sylvain Leblanc of the Royal Military College in Canada — has designed a way to detect unusual activity in a car’s computer system. The trio is using machine learning, noting that anything out of the norm within an operating system may signal a cyber attack. The goal is to take this information and create the tools needed to address these security threats to connected vehicles before they turn into action.

In Washington, D.C., American University’s computer science department is currently focused on what they are calling “vulnerability management.” This entails “detecting attack during normal system operations.” We’ve discussed many ways cars are now more open to outside interference by the mere fact that there are a vast number of ports of entry into a car’s operating system. Whether it’s through telematics or that cool new infotainment experience, there are all of these little ways your high tech vehicle can be infiltrated. By paying attention to changes in normal computer behavior, the hope is that this will immediately signal a need for cybersecurity experts to step in and, at the very least, investigate the issues arising in the automobile.

As Professor Japkowicz and team engage two machine learning techniques to gain insight into activity that is out of the norm — Long Short-Term Memory and Gated Recurrent Units — they hope to provide the automotive industry with the tools to battle and evade such attacks. Newer cars are far more vulnerable than older models to any sort of cyber invasion with more computerized capability than ever before. From the simple USB port on your dash to the intricate framework of autonomous driving, the number of ways to enter from the outside has heightened exponentially and the challenge is to now shut each and every one of those areas from outside interference. That, in itself, is a lofty task especially as more cars coming off the assembly line are outfitted with more advancements, more bells and whistles, and more connectivity than ever before.

Take for example the recent slew of Tesla thefts in England. Thieves have been hacking into key fobs to unlock a Tesla, start it up and drive it away. Tesla reacted by offering an optional “PIN to Drive” feature, helping owners disable “Passive Entry,” and making a signal blocking pouch in which to keep your key fob available. This has helped, but those owners who don’t use any of the tips may find themselves losing their vehicle thanks to those who are technically proficient enough to become part of the New Age of Car Thieves.

Sidebar: “What if…?”

Although virtually accessing self-driving cars is the biggest concern of both vehicle manufacturers and the tech industry — not to mention what all of this means to your insurance premiums — this got us to thinking…

Imagine you’re slowing down to a stop for a pedestrian crossing the street when the front door lock is jimmied — either manually or digitally— and someone slips inside to take control. Autonomous car theft is consistently viewed from a high-tech, cyber-hacking standpoint, but what about good, old-fashioned carjacking? Without a driver, some skill with either an actual lock-pick or rudimentary hacking knowledge may very well turn that fully autonomous dream into a vintage nightmare. Hmm…

Tracking the hacker? Not so easy

hackers for autonomous vehicle theft

While more than 90% of all new cars have a black box installed to assist with determining issues when a crash happens, finding the hacker who cracked your automobile’s computer code is very difficult. They range from networks of people who work together to solo infiltrators writing self-erasing code, fake web addresses and more. Since there is no physical evidence of the break-in — only virtual — investigations can sometimes take years before the perpetrators are found. So, where does that leave you comfortably sitting in the backseat of your self-driving vehicle that is suddenly taken over and driven to…? Well, you get the point.

For now, all of these incidences are virtually nonexistent, but tomorrow? As someone once said, “Tomorrow is another day.” However, that “day” is fast approaching.


, , , , , , , , ,

AutoMobility LA 2017: Revealing Car Exhibits (Day 3)

Part Three of our four-part series recapping the summit’s four days of exhibits, announcements and interactivity

Car exhibits open at AutoMobility Day 3

Day 3 of AutoMobility LA kicks off with vintage car exhibits

Beginning in the early morning with a walk down vehicle memory lane, AutoMobility LA’s third day focused on auto reveals, announcements and interactivity. Car exhibits opened and the experience provided by each automaker was meant to engage you in the automobiles they were touting as intensely and immersive as possible. Several unique partnerships were unveiled, and gamification and VR played heavily throughout.

Nissan embraces entertainment with car exhibits

Nissan’s interactive Star Wars display

Nissan offered a VR simulator in which summit goers could “repair” a Bb-8 unit from Star Wars. Chevrolet had a stand where you could choose one of its models to design your own car.

car exhibits display immersive experiences

Acura provides virtual reality racing while Nissan entertains with Robert Delong

How each new announcement was made at each of the car exhibits reflected the brand — Lincoln presented its new Nautilus crossover after treating all those who attended to a live set of a piano accompanied torch singer, champagne, imported cheeses and personalized chocolate truffles with the Lincoln logo emblazoned on them. Lexus projected a sleek trailer featuring actress Minnie Driver — the voice of Lexus. Nissan’s brief but energetic reveal of the new Kicks, a $19,000 two-tone Rogue-like five-seater, was preceded by the electronic musical stylings of singer Robert Delong performing live. And Hyundai had a DJ mixing and spinning in the minutes leading up to its media event.

The key to everything on this third day was customer service. Concierge and personal service played huge roles with many of the car exhibits and brand goals, as did alternative fuels and more connectivity as ways to engage consumers. In some cases, returning to the roots that built a brand to ensure fans knew they were not forgotten was important and as Day 3 began, it was clear the auto industry was preparing for and proactively addressing the future.

Intel entertains autonomy

car exhibits highlight autonomy

Brian Krzanich, CEO, Intel, discusses the company’s driverless future

AutoMobility LA began the morning with a talk by Intel CEO, Brian Krzanich. Autonomous is a huge investment for intel and they have set up a variety of partnerships as they look at the future of autonomy and how this not only affects how people drive, but cities themselves. To that end, they will be rolling out 100 driverless cars as a test, partnering with Waymo to target Level 4 & 5 driving. In this new normal for Intel, the company has also partnered with Warner Bros. (WB) Entertainment to bring an even more immersive experience to autonomous vehicles. Riders will be able to have their windshields and surrounding windows filled with trailers, movies, animations, and basically be surrounded by WB content. One of the 100 cars being tested is earmarked as a pilot for this new partnership.

Mercedes brings Formula One to the streets

Mercedes’ announcement included a big move forward in its goal to develop for alternative fuels and staying true to its automotive legacy. First, the establishment of EQ, the new brand that stands for intelligent electric mobility — a comprehensive electric eco-system that focuses as much on personal service as it does technology, innovations and products. The goal is to produce electric cars by 2022 and EQ SUVs will be made in the company’s Tuscaloosa, Alabama, plant, creating over 600 new jobs in the area.

car exhibits present unique transport

The Mercedes-AMG Project One, the world’s first street legal Formula One

Tobias Moers, Chief Executive of Mercedes-AMG, announced the first street legal Formula One car in history with the AMG Project One.

Porsche returns to high-performance basics

car exhibits unveil a return to roots

The 2018 Porsche 911 T Carrera

Perhaps one of the most alluring car exhibits exuding an air of immense anticipation was that of iconic carmaker, Porsche, set up in its own limited access showroom. In addition to a new 718 GTS, Panamera hybrid and Cayenne, Porsche presented its new 911 T Carrera. It is a return to its down and dirty high-performance, powerful roots.

Mazda goes upscale

Mazda seeks to enter luxury with its car exhibits

The Mazda Vision Coupe and updated 2018 Mazda6

Mazda has long been perceived as an economical brand that offers good rebates. At AutoMobility LA, the carmaker announced both its Vision Coupe — a sleeker version of its iconic RX7 — and the Mazda6, two upscale models meant to move the company into the luxury market. It is an attempt to change perception and allow them to compete with the higher end brands.

Nissan: May the Force awaken the Last Jedi

car exhibits entertain

Nissan presents Kicks, its two-tone take on the iconic Rogue, and a partnership with Star Wars

As car exhibits go, Nissan’s was among the most interactive. The two-tone Kicks wasn’t the only big announcement from the Japanese carmaker. Not only did Nissan offer VR droid repair simulations, but announced its partnership with Star Wars: The Last Jedi by presenting various refurbished car models in the Star Wars vein.

car exhibits from Nissan's Star Wars partnership

Nissan’s souped-up Star Wars themed vehicles

A promise of even more information on this new partnership at the fourth and final day of AutoMobility LA was made by Nissan.

Car exhibits reflect entertainment partnerships 

Lexus also discussed its much touted partnership with Marvel’s Black Panther franchise, sharing its limited edition LC featuring the Black Panther logo on the hood. Film, television, and gaming intersected loudly at the summit — another feature employed by just about every one of the car exhibits in some form or another — and were clear indications of how instrumental and fertile these partnerships are to the continued success and appeal of the auto industry.

, , , , , , , , , , , , , , , , , ,

AutoMobility LA 2017: Securing Our Autonomous Future (Day 1)

Part One of our four-part series recapping the summit’s four days of exhibits, announcements and interactivity 

The 2017 Los Angeles Auto Show kicked off on Friday, December 1st, promising some stellar exhibits, surprise announcements and an immersive motor vehicle experience for all who attend. The show is celebrating its 110th year and has expanded from just 99 cars in Morley’s Skating Rink in Ocean Park, Los Angeles, back in 1907 to taking over virtually every available space of the Los Angeles Convention Center in the city’s Downtown area. However, for those interested in learning even more about what’s coming in the future of cars and the automotive industry as a whole, there is a show before the show in the form of AutoMobility LA, four days of announcements, thought provoking conferences, and first looks at innovative cars.

AutoMobility LA sets the stage from Day 1

AutoMobility LA entered its second year as it took place from November 27 – 30. It was a mecca for those curious about the intersection of technology and cars as well as hoping for a peek at what auto companies have up their sleeves for the future. Over the course of four days, AutoMobility LA covered everything from concerns about cyberhacking in the connected car space to Porsche’s return to its roots and more while also acting as an intensive primer on where the industry is headed.

Secure Mobility jumpstarts discussion

AutoMobility LA 2017’s first day was kicked off by the day long Secure Mobility Summit. This informational conference focused on cyber-security for connected cars and, most specifically, driverless vehicles in general.

Presented by Suits & Spooks founder, Jeffrey Carr, in conjunction with Car and Driver and other sponsors, Secure Mobility included several featured speakers and highlighted five key takeaways:

1 — Get In Front of the Breach

Every car maker is aware that as automobiles become more sophisticated, these machines are more vulnerable to tech savvy interference. Waiting for something to happen rather than being getting in front of it will only lead to worse problems down the line.

2 — Willingness to Invest

Comprehensive security of a car’s technology is expensive. But as each and every speaker said, the cost of NOT doing anything far outweighs being proactive in staving off a cyber-attack before it happens.

3 — Share Information

Auto manufacturers need to trust their competitors with any information that will lead to keeping everyone safe in their cars. Holding back intel regarding hacking of a system or a discovery on how to stop an attack can lead to the entire industry being brought down.

4 — Assume your car is always connected

Any car produced after a certain year is connected to the web in some way. Hands free talking through your dashboard, an insurance fob plugged into your onboard diagnostics (OBD) port, and that cool infotainment system connect you to the outside world. Your vehicle’s ability to communicate with the internet of things (IOT) means a constantly open digital doorway. However, there’s good news.

5 — Vehicle cyber attacks are DIFFICULT

Some basic internal configurations are shared across all vehicles, true, but ultimately, all those things that define a model’s specific operating system as being different from another model are what also makes them difficult to crack inside. It means if you really want to break into a system to bring a vehicle down, you need to learn everything there is to know about that particular auto’s OS and determine its weaknesses. It’s not so easy to take a car offline, although while a presenter was talking, his cohort was able to hack a navigation system as a demonstration on how quickly and easily GPS’s can be compromised with the right information and tools. Individual cars are generally well-protected, and as they continue to innovate, so does their security. However, commercial fleets share the same OS across a broad range of vehicles, and concern is high on how to ensure thousands of trucks and vans aren’t attacked all at once.

Shared cyber-security responsibility

Most of the ownness on protecting connected cars is on the auto industry, but as consumers of motor vehicles, there must be something we can do. When Sanchi Jayarami, Chief of Investment and Engagement at the National Security Division (NSD) — part of the Department of Justice (DOJ) — was asked what vehicle owners can do to secure their automobiles, she pointed to the risks and concerns discussed and said, “My parents said, ‘This is why we don’t use an ATM and we go into the bank to talk to a person.’ They’re not wrong. It’s important people read about what their car can do. How it works.”

Sanchi added, “I know it’s hard because of all the small print, but they’ve got to do it so they know how their information is being used and how they can protect it. And it’s important for companies to educate their consumers.”

As the first day came to an end, it was clear that the brilliance of the driverless cars and connected vehicles in general is as much a benefit as it is a concern. Both automakers and consumers need to be ready for what the future holds with more innovation, higher level technology, the reach of artificial intelligence (AI) in motor vehicles and the new players coming into the space. Each needs to be responsible for protecting and knowing their vehicles in order to be ready to do whatever is necessary to keep themselves and those they encounter on the road safe.

, , , , , , , , , , ,

AI in Self-Driving Cars: Deep Learning Deciding Life or Death

AI in self-driving cars

Creating a safer, less congested and ecologically friendly world is the main goal of driverless vehicles, and artificial intelligence or AI in self-driving cars is the brain behind making these modern marvels work. In order to get these automobiles on the road, the global community is doing its best to support autonomous car development. The current U.S. Department of Transportation Secretary, Elaine Chao, held a press conference at the University of Michigan’s, MCity in late September to announce the simplification of restrictions on autonomous vehicle R&D and testing. A bill addressing how this will be done is currently making its way through the U.S. Senate and House. Europe is already testing autonomous shuttles in some cities — driverless public transport has long been seen as a natural fit for the technology — and companies like nuTonomy in Singapore and Uber in various cities of the United States have been trying out autonomous taxis and ride-hailing options on public roads for a few years now. The inputs and outputs AI has to access just for the normal operation of a vehicle on everyday roads are vast and complex but add to that the unique decisions of life and death put on the neural networks, and innovators are seeking ways to make those less polarizing. After all, as much as many look forward to a day when the driverless car makes traffic jams and automobile collisions a thing of the past, one issue keeps coming to the surface: what happens when an unmanned vehicle is faced with whom to save in an impending accident?

The Spock conundrum of logic vs. emotion

left and right brain assimilating the info provided for AI

Left Brain (logic) vs. Right Brain (emotion)

Up until recently — and in some cases still — self-driving cars have been seen as a thing of fantasy and science fiction. To that point, sci-fi stories have long used logic to make choices that emotion may very well undermine. Logic-based life-forms and sentient beings tend to react in ways that are seemingly counter to humans, but often end up being about salvation rather than destruction. The purpose, obviously, is to show the human side of a machine or emotionless entity in a way that gets the audience to embrace the character. However, when you break down those decisions, you realize that those choices are logical, even as they tug at our emotions.

In the classic film, Star Trek: The Wrath of Khan, as the being synonymous with bridging the gap between emotion and logic sacrifices himself for the crew, Spock opines, “The needs of the many outweigh the needs of the few… or the one.” It is a logical choice when faced with such a situation — one or few lives as opposed to masses. Arnold Schwarzenegger’s reformed terminator in Terminator 2: Judgement Day concluding that sacrificing himself to save a race he was once programmed to destroy comes from the deep learning of his neural networks. He has come to understand and care about them, and sacrifice to save is the logical choice so that the needs of the many are met. K-2SO’s going out in a hail of laser cannon fire in Rogue One: A Star Wars Story so that Jyn and Cassian get the plans to the Death Star, thereby saving billions of lives in the process is, also, a logical decision made by another sentient being. And the doe-eyed, heartstring-pulling WALL-E hitching an unscheduled ride on the spaceship that has taken his beloved and suddenly catatonic EVE (EE-vah) only to discover the entire race of the planet he’s been cleaning up is being relegated to unhealthy, fat and forever lost in space drones. His giving himself over to losing his learned humanity to save them all is, once again, the logical choice. One trash-gathering robot v. the whole human race? No question.

Therefore, programming the AI in self-driving cars to react in a way that serves the greater good makes sense, right? However, road collisions are frequently one-to-one situations. They rarely enter the “needs of the many outweigh the needs of the few” category, at least on the surface. It’s that grey area that adds an extra dimension to coming to a conclusion that makes the most sense. For human drivers, this is a moral dilemma that is only realized after much thought — weighing the cause and effect — and consideration — the emotional burden of a decision that leads to possible tragedy. The “moral” only comes into play when a person is operating a vehicle, but when artificial intelligence in self-driving cars accesses digital programming to determine what to do, deep learning comes in. Understood. Except, what is deep learning?

Teaching machines how to learn

deep learning in a neural network

Deep learning basically takes massive amounts of data, layering it upon each other to build conclusions that lead to a human-like recognition of what something as abstract as an image or sound actually is for a machine. The “deep” comes from how, over time, the levels of information gathered and how it experiences this data leads to it learning more about what it is gathering, allowing the AI to correct itself and be better at recognizing and, therefore, reacting to the input appropriately. Correcting its own mistakes, just like you and me.

Heady stuff, right? Well, these are exceptional amounts of data being assimilated, and an amazing capability that is pushing automakers and tech giants closer to creating truly effective driverless solutions. HOWEVER, it still doesn’t answer the question: what happens when a machine is faced with who lives and who dies on the road?

a car stopping for a pedestrian

Here’s the dilemma—and one we discussed in our Child Safety article: You’re in your self-driving car, barrelling happily down the street when a pedestrian runs out. On one side is oncoming traffic, the other a sheer cliff. What do you do? If you, a human, were driving, many people say they would go off the cliff, hoping to miss the pedestrian AND the many in oncoming traffic. But, if there is no one behind your wheel, if you’re not manning it, the AI in self-driving cars kicks in and makes the decision for you. And per automakers, that decision is to save the passenger in the vehicle, not the people on the road. What would be a big “Whoa!” moment that you will live with for the rest of your life is not so for something like NVIDIA’s Drive PX. Because AI in self-driving cars doesn’t have the ability to emotionalize things, it will never wake-up in the middle of the night, sweating about the choice it makes forever. And because of this lack of a moral dilemma or ethical consideration, AI in self-driving cars is basing the decision on algorithms and probabilities, patterns of recognition that feed deep learning, not emotion. And how is all of that processing getting artificial intelligence to the right decision?

Good question. Because, when you think of it, that whole “the needs of the many outweigh the needs of the few” is flipped in a driverless world. It becomes, “the needs of the ones in the car outweigh the needs of the bunches out there on the street.” It’s a choice, sure, but is it a truly logical one?

The reasoning power of AI in self-driving cars

How AI in self-driving cars may react and see things

Before we go into what kind of information artificial intelligence is taking in to help it decide who lives and who dies, perhaps we can play a bit with what we, the humans, do with the inputs we’re given.

Massachusetts Institute of Technology (MIT) worked on a self-driving car project that included considering this dilemma and created a site that allows people to test what they would do in these potential crash situations. As you move through the different scenarios, it gets harder and harder to decide what is “right,” even to the point that you wonder if there is any “right” in these instances. And if we, as thinking humans, can’t figure out what’s the right thing to do, how does AI in self-driving cars? By taking the emotional connection to the problem out of the equation and using algorithms feeding deep learning, autonomous vehicles are able to do what they need to do in order to get from Point A to Point B, efficiently and seamlessly. Collateral damage may very well be something with which to contend, but, as automakers have made clear, the technology behind each driverless car is created to manage slowing down, averting, braking in enough time to avoid loss of life and catastrophic auto accidents.

However, it’s still a rather unique “moral” problem. From ethicist Philippa Foot to philosopher Judith Jarvis Thomson and beyond, figuring out how to address this ethical issue has been a challenge. Everything about the Trolley Dilemma is based upon a series of factors you may never face. The world is moving forward with autonomous vehicles and the reality of actually having the AI in self-driving cars use its neural network to make a decision that is beyond passenger control is on the horizon. An example of a supplier of this technology is Drive PX, Jen-Hsun Huang’s company, NVIDIA’s solution offering small and large options for making a vehicle’s brain self-driving. That something like this is now available makes understanding the mechanism and reasoning behind such choices — deep learning — more pressing.

algorithm on AI in Self-Driving

Algorithm image by Docurbs via Wikimedia Commons

Machine learning, the algorithm way

AI in self-driving cars becomes smarter thanks to algorithms. But what exactly are these? And how do they learn or contribute to the “smartness” of your car?

An algorithm is comprised of inputs that prompt specific outputs. It’s a series of bits of information fed into a centralized mechanical brain that tell that brain how to take that INPUT and create actionable OUTPUT to initiate an appropriate response. An algorithm is likened to a recipe—the ingredients being the inputs and the meal the outputs. It’s a helpful tool in the world of machine learning, and in the case of AI in self-driving cars, a huge influence on creating a safer, seamless experience.

How that information used to create these algorithms is truly assimilated and accessed by human beings is a concern. An example given in the article “Why AI Makes It Hard to Prove That Self-Driving Cars Are Safe” by Andrew Silver poses the worry over what is being considered and what isn’t:  “Imagine if you test drive your self-driving car and want it to learn how to avoid pedestrians. So you have people in orange safety shirts stand around and you let the car loose. It might be training to recognize hands, arms, and legs—or maybe it’s training to recognize an orange shirt.”

Because algorithms use statistics to define results, they have the potential to put in or take out specific information to come to a conclusion. This incompleteness leads to a familiarity that, in instances like this, can breed danger, creating limitations. That type of rote learning can mean something as simple as getting to know a certain area of town really well while not understanding other parts at all. This would make successfully navigating those alternative roads, street signs and pedestrian interactions virtually impossible.

Difference between Traditional Modeling and Machine Learning. Image from ZEISS International

There are two unique types of algorithms used to make AI in self-driving cars effective: Computer Vision and Machine Learning. Computer Vision algorithm is the more traditional form that uses a cascading sort of learning process with encoded programming that leads to a predicted result. The newer, more innovative and precise Machine Learning algorithm, also known as deep neural networks, goes beyond basic codes and uses sample data to “learn” and assume results on what it has yet to experience, thereby broadening its parameters of output. In the case of deep learning, the data accumulated to feed those decisions go into something called a “black box.” It holds all of that information so it can be accessed and used by the machine’s brain. However, the actual process of the inputs leading to the outputs is so intensive, comprehending what led to that decision is beyond human thought. This means that should the system react incorrectly, it’s virtually impossible for a person to take all that’s been gathered in the black box and determine what caused the wrong decision. And if they can’t figure that out, they won’t be able to fix the actual process that led to that conclusion so it won’t happen again.

AI in self-driving car as the vehicle sees it

Introduce into that deciding who on the street to save in case of an altercation. If the mechanical brain of the autonomous vehicle has input probabilities that only have a finite number of outcomes and finds itself faced with one that has never been encountered before, how does it make the decision? It pulls from what it knows, adjusting to the situation as best it can with the information it’s been given. Much like humans, actually, however with a more logical and detached view that allows that car to react in a way that protects its passenger above all else. Simple, straightforward and unfailing. Because that whole idea of “the needs of the many outweigh the needs of the few” has a lot of deep layers. Whose many? What few? And how many are affected if that few — or one — is lost? It goes on and on and on until, honestly, some sort of stance needs to be made and just as the MIT project shows, figuring out a clean, clear choice when there are so many variables involved is virtually impossible. For a machine, picking a lane comes without any baggage. But for the human at the effect of that decision? It is a weight far greater than anything a machine can comprehend. The capacity for understanding flips — people know there is sacrifice and know the outcome of that brings with it pain and confusion that must be bore. Machines only act and as much as they may or may not learn from their action, they will never feel the enormity of it.

An enduring question

It is a given that humans are fallible. There is enough data to warrant a look at the high cost of human error when it comes to driving motor cars. Something Silicon Valley is also learning through its testing — from companies like Uber to Waymo and beyond — is that people are also detrimental to a self-driving car project — the bulk of autonomous car accidents are due to human intervention. And studies have shown that machines react much faster than a man or woman ever could. But if the AI in self-driving cars isn’t sure what it’s reacting to or how it’s supposed to react, is it even possible for it to take action at all, let alone appropriate action?

where AI in self-driving cars is taking shape

A view of Silicon Valley at dusk

It’s a unique conundrum, surely, and there are no easy answers in any of this. And, so, as these various Silicon Valley giants and automakers discuss, confer, and keep coming up with such technologies as Jen-Hsun Huang’s Drive PX, deep learning of algorithms, and neural networks in general, the technology gets smarter, but the questions get harder.

, , , , , , , , , , , , , , , , ,