No Speed Limit Autobahn and AI Autonomous Cars

By Lance Eliot, the AI Trends Insider

The joys of driving on the German autobahn!

Nearly two-thirds of the 8,100 miles of the autobahn system allow unrestricted speeds. This means that there is no speed limit. Go as fast as your heart desires. Go as fast are your car can go. Go as fast as your fears will let you proceed. I mention the fears part of this driving experience because there are some drivers that frankly are not so keen on driving at super high speeds and they either fear the consequences or at times have panic attacks about it.

I lived in Germany for a year and personally morphed through several mental and emotional states about the autobahn.

At first, I was quite excited about being able to drive at blazing speeds. Having come from the United States for my yearlong visit to Germany, I was well acquainted with the fact that mighty speeds were pretty much verboten in my home country. Sure, I had occasionally had a lead foot as I drove from Southern California to the Las Vegas strip in Nevada, everyone drives fast through the desert areas that are merely an impediment to getting to the tables in Vegas and laying out some bets. But this over-the-top speeding came with the chance that you would get stopped by the police and it was as much a bet to get a ticket and a hefty speeding fine as were the odds of losing your savings while gambling in Vegas.

Believe it or not, I gradually kind of got used to the fast driving of the autobahn. Interestingly, I also discovered that I had my own threshold of how fast I was willing to go. I also discovered that there seems to always be someone wanting to go faster than you, no matter how fast you might think you are going. I would shake my head in disdain at cars that came up to my bumper, while each of us were both driving already at hyper speeds, and dutifully would try to get out of their way. It was surprising to see them scoot away from me and zoom ahead, given that I was already going at such a high speed, and I tried to guess how much faster they must be going to be such a blur.

Free Travel For Free People

The colleagues and friends that I had in Germany would tell me that the autobahn unrestricted speed limit was essentially a birthright. The oft repeated slogan is “Free travel for free people.” This seemed to therefore intertwine with their culture and their psyche. You dare not try to take away their unrestricted speed. You would need to pry their dead cold hands from their steering wheel to try and do so, which I mention as a kind of nod to how some Americans perceive their rights related to guns.

One aspect about the autobahn that I admired was how well kept it seemed to be. One would hope that when trying to use a road for such high speeds that you would keep the road in decent shape. The dangers from potholes and other roadway surface issues are greatly magnified when you are traveling at tremendous speeds. The car will react instantaneously and the human driver might not be able to react as quickly, thus, any roadway imperfections can have a rather daunting result.

There were an estimated 400 people killed in autobahn car incidents last year. This number of course is unfortunate. When put in light though of the number of miles driven, the per capita basis is actually not statistically abysmal. With a population of around 83 million, and the number of cars in Germany estimated at 46 million, there are perhaps a total of 350B to possibly 385B miles traveled (often referred to as VMT, Vehicle Miles Traveled). The average number of miles driven per car is estimated around 8,300 miles.

When you take a look at charts of deaths per 100,000 people due to driving incidents, Germany is around 4.3, which is actually relatively low, and a lot less than the estimated 10.9 in the United States. The point of these numbers and statistics is that it doesn’t seem that the unrestricted speeds of the autobahn are somehow leading to a vastly larger number of car related deaths. I’ve not seen any published numbers that are able to make the case that the autobahn is a magnet for car related deaths or otherwise pushes up the likelihoods of car deaths.

That being said, there are certainly some quite well-known consequences of the high speeds on the autobahn. Few would dispute the idea that when there is a car accident on the autobahn, the odds are that the results can be fairly horrifying. If a car spins out of control at hyper speed, it is going to cause some armor busting damage to anything nearby. Roadway railings will get bent out of shape. Other cars nearby are going to become pinball targets. And so on.

Another aspect is that there is a solid chance that when a car accident occurs, it can become a cascading one. Other cars that come upon an incident might not have sufficient time to slow down or otherwise react. They then can be caught up into the incident. A series of cars piling up can happen readily. The closest I’ve experienced in the United States has involved cars driving in the fog that cascade into each other when a car accident happens, and also sometimes I’ve seen the same happen when there is ice or snow on the roadways.

My colleagues and friends there in Germany emphasized to me that it was the responsibility of each driver to get out of the roadway as quickly as possible whenever having any car troubles. I recall one time that I was driving on the autobahn and had a colleague with me, and I saw dark smoke billowing up ahead. I started to slow down. My colleague looked at me with a puzzled expression and asked why I was slowing down. I told him that I wanted to be careful as we approached seemingly an adverse car accident up ahead (I assumed a car was likely on fire after a car crash of some kind).

He got irritated at me and insisted I proceed along at full speed. Why, I asked him? He assured me that the German driver that was up ahead would certainly get out of the way, as prescribed by law and by the accepted credo of drivers there. If I slowed down it would only likely cause me to get into a car accident because some other car from behind me would get confused about why I was slowing down. I admit this logic was a bit beyond my usual way of thinking and I wondered whether an injured or possibly dead driver in the perhaps car accident up ahead would really have been able to get off the road or not.

In any case, I did survive the autobahn experience while in Germany for the year and both enjoyed driving on the infamous autobahn and at times had some trepidations about it. One of those love-hate relationships, I suppose.

Some Interesting Facets About The Autobahn

One aspect that was vividly brought to my attention was that the autobahn was a source of pride for the country. I don’t know of any other place on earth that has this unrestricted speed limit for such a large geographical area of thousands of miles and that has somehow made it all work. Most of the locals did not particularly bring up the autobahn as a topic, since they take it for granted. It is there, it exists, it works, and they are happy with it.

They did point out that it acts as a tourist attraction. I am used to places like Disneyland in Southern California as being a tourist attraction, or perhaps Hollywood where you can see the movie stars. Thinking of the autobahn as a tourist destination was intriguing. I did meet a number of tourists that told me they might have not otherwise come to visit Germany except that they especially wanted to drive all out on the autobahn.

Besides the pride aspects and the tourist venue, there is also the factor that Germany makes cars. There is an image of German made cars that can zip along on the autobahn. This imagery is perhaps a significant element for why people worldwide desire to purchase German made cars. In that sense, you could say that the autobahn is intricately bound into the German economy. Car production and the selling of German cars might not do well if the autobahn did not have the unrestricted speed limit and the available branding that goes with it.

This discussion of the autobahn would not be complete if I didn’t bring up an ongoing debate and concern within Germany about the unrestricted speed aspects. There is a storm of sorts brewing in Germany about the potential for a speed limit on the autobahn (note, as mentioned earlier, there are stretches that already do have speed limits, and so the debate is about potentially covering more or all of the autobahn with speed limits).

The debate is focused on pollution, rather than car safety aspects per se.

Turns out that the cars zooming along on the autobahn are pouring out carbon emissions. This butts up against the German led efforts to fight pollution and also their desire to be a leader in the climate change arena. The German Federal Environment Agency has suggested that by setting the speed limit to 75 miles per hour, the result could known down carbon emissions by cars on the autobahn, decreasing the foul output by perhaps 9% per year. Generally, the floated speed limit number is around 75 to 80 miles per hour.

Geschwindigkeitsbeschränkung! That’s the lengthy German word meaning maximum speed limit. It’s a word that for some Germans is a foul one, at least with respect to the autobahn. As mentioned earlier, the unrestricted speed is a key source of pride, it attracts tourists, it ties into the economy, it has merged into the culture and mindset of the people, and so anyone trying to mess with it is going to find themselves confronting quite a fierce resistance.

Some of my friends and colleagues felt that driving at 75 or 80 miles per hour is a type of insult. It is beneath anyone of proper driving ability. It also underuses the capability of the car. Why would you have a finely made precision car that can go much faster and yet limit it to being no more than a slowpoke on the roadways? They are sympathetic to the noxious emissions factor and regret that pollution is a byproduct, but they are not convinced that the speed limit imposition is the means to deal with the problem.

It’s an internal political and public battle that can tear apart the best of friends and spark rather heated discussions. Seems like everyone has a diehard opinion on the matter.

AI Autonomous Cars And The Autobahn

What does this have to do with AI self-driving cars?

At the Cybernetic AI Self-Driving Car Institute, we are developing AI software for self-driving cars. One interesting question is whether or not AI self-driving cars might be driving on the autobahn at unrestricted speeds, and also whether the advent of AI self-driving cars might play into the future of the autobahn.

Allow me to elaborate.

I’d like to first clarify and introduce the notion that there are varying levels of AI self-driving cars. The topmost level is considered Level 5. A Level 5 self-driving car is one that is being driven by the AI and there is no human driver involved. For the design of Level 5 self-driving cars, the auto makers are even removing the gas pedal, brake pedal, and steering wheel, since those are contraptions used by human drivers. The Level 5 self-driving car is not being driven by a human and nor is there an expectation that a human driver will be present in the self-driving car. It’s all on the shoulders of the AI to drive the car.

For self-driving cars less than a Level 5, there must be a human driver present in the car. The human driver is currently considered the responsible party for the acts of the car. The AI and the human driver are co-sharing the driving task. In spite of this co-sharing, the human is supposed to remain fully immersed into the driving task and be ready at all times to perform the driving task. I’ve repeatedly warned about the dangers of this co-sharing arrangement and predicted it will produce many untoward results.

For my overall framework about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/framework-ai-self-driving-driverless-cars-big-picture/

For the levels of self-driving cars, see my article: https://aitrends.com/selfdrivingcars/richter-scale-levels-self-driving-cars/

For why AI Level 5 self-driving cars are like a moonshot, see my article: https://aitrends.com/selfdrivingcars/self-driving-car-mother-ai-projects-moonshot/

For the dangers of co-sharing the driving task, see my article: https://aitrends.com/selfdrivingcars/human-back-up-drivers-for-ai-self-driving-cars/

Let’s focus herein on the true Level 5 self-driving car. Much of the comments apply to the less than Level 5 self-driving cars too, but the fully autonomous AI self-driving car will receive the most attention in this discussion.

Here’s the usual steps involved in the AI driving task:

•         Sensor data collection and interpretation
•         Sensor fusion
•         Virtual world model updating
•         AI action planning
•         Car controls command issuance

Another key aspect of AI self-driving cars is that they will be driving on our roadways in the midst of human driven cars too. There are some pundits of AI self-driving cars that continually refer to a utopian world in which there are only AI self-driving cars on the public roads. Currently there are about 250+ million conventional cars in the United States alone, and those cars are not going to magically disappear or become true Level 5 AI self-driving cars overnight.

Indeed, the use of human driven cars will last for many years, likely many decades, and the advent of AI self-driving cars will occur while there are still human driven cars on the roads. This is a crucial point since this means that the AI of self-driving cars needs to be able to contend with not just other AI self-driving cars, but also contend with human driven cars. It is easy to envision a simplistic and rather unrealistic world in which all AI self-driving cars are politely interacting with each other and being civil about roadway interactions. That’s not what is going to be happening for the foreseeable future. AI self-driving cars and human driven cars will need to be able to cope with each other.

For my article about the grand convergence that has led us to this moment in time, see: https://aitrends.com/selfdrivingcars/grand-convergence-explains-rise-self-driving-cars/

See my article about the ethical dilemmas facing AI self-driving cars: https://aitrends.com/selfdrivingcars/ethically-ambiguous-self-driving-cars/

For potential regulations about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/assessing-federal-regulations-self-driving-cars-house-bill-passed/

For my predictions about AI self-driving cars for the 2020s, 2030s, and 2040s, see my article: https://aitrends.com/selfdrivingcars/gen-z-and-the-fate-of-ai-self-driving-cars/

Returning to the topic of the autobahn, let’s consider the ways in which AI self-driving cars might come to play regarding the autobahn matter at-hand.

Let’s start with the matter that has got many people up in arms about potentially putting a speed limit on the autobahn, namely the carbon emissions of cars that ride on the autobahn.

For AI self-driving cars, it is likely that most or nearly all AI self-driving cars will be EV’s (Electrical Vehicles), and as such the tailpipe emissions issues goes away. Some argue that you need to calculate the emissions due to power plants that will produce the electrical energy needed for EVs, and perhaps add that to a per capita sense of “emissions” for EVs, but I’m not going to go that route herein. Let’s generally say that the carbon emissions coming out of an EV car is zero.

Why would most or nearly all AI self-driving cars be EV’s? Partially due to the aspect that the AI self-driving car is an electrical power guzzler of sorts, needing to use electrical power for the multitude of added sensors, the multitude of added computer processors, and other add-ons that make an AI self-driving car a functioning AI self-driving car. You can generate the power via conventional means and use a gasoline or diesel fueled engine, but it seems like it will be easier to go with an EV. There is also usually a gigantic battery in an EV, while the battery for a conventional car is typically puny.

Plus, there seems to be a hope that the advent of AI self-driving cars will persuade people to switch from gas guzzlers to EVs. Besides the EV being a natural fit due to the ability to generate the needed electrical power and have large-scale batteries on-board the car, the thinking is that if people want the AI self-driving car aspects, they might be more willing to switch to an EV if that’s the type of vehicle being used for AI self-driving capabilities. This would help out the environment and reduce too a dependency on gasoline.

In a manner of speaking, the AI self-driving aspects might be the icing on the cake to get people over to EVs. Right now, there are only about 1% of all cars in the United States that are EVs, and so it is a tiny portion of the car population. Assuming that we aren’t going to try and retrofit existing cars to be AI self-driving cars, it means people are going to be buying new cars. If they are going to be buying new cars, an incentive to buy an EV would be the AI self-driving capability.

For more about EV’s and AI self-driving cars, see my article:  https://www.aitrends.com/ai-insider/power-consumption-vital-for-ai-self-driving-cars/

For why retrofitting existing cars to be AI self-driving cars is unlikely, see my article: https://www.aitrends.com/selfdrivingcars/kits-and-ai-self-driving-cars/

For my article about the affordability of AI self-driving cars, see: https://www.aitrends.com/selfdrivingcars/affordability-of-ai-self-driving-cars/

For my article aspects about mass transit and AI self-driving cars, see: https://www.aitrends.com/selfdrivingcars/personal-rapid-transit-prt-and-ai-self-driving-cars/

For the timetable of AI self-driving cars, see my predictions: https://www.aitrends.com/selfdrivingcars/gen-z-and-the-fate-of-ai-self-driving-cars/

Carbon Emissions Considerations

The key point is that if we are going toward AI self-driving cars, eventually, inexorably, and if those AI self-driving cars are mainly EV’s, it would seem to “solve” the autobahn qualm about the carbon emissions.

For Germany, they are faced with having to ascertain whether they believe that AI self-driving cars are coming or not, and if so, and assuming the use of EV’s will then predominate, they could wait out the existing carbon emissions issue and be able to assuage it during the switchover to AI self-driving cars.

The question then becomes how much pollution they could have otherwise reduced in the interim by setting speed limits and does that amount of emissions reduction “payoff” for the consternation of restricting the speed limits of the autobahn until the arrival of AI self-driving cars.

Notice that this is a different approach to the matter. For most debates, the usual argument is to permanently set speed limits. In the case of this alternative argument, the question becomes the interim period rather than a forever kind of timeline.

For those that are pressing for the carbon emissions reduction, any furtherance of emissions is bad, and so they would say that the speed limit restrictions could be temporary, and this should satisfy the “no speed limits” proponents. Those proponents just need to keep their eye on the emergence of AI self-driving cars and once that happens, presumably the speed limit goes back to unrestricted.

For those that are vocal about keeping unrestricted speed limits, they would say that the lack of a speed limit should continue unabated and that the emissions contingent just needs to wait for the advent of AI self-driving cars. The problem of foul emissions, in a manner of thinking, will be going to solve itself, doing so without having to mess with the existing approaches such as speed limit aspects.

Get out your crystal ball because the interim period could either be short or long, depending upon how fast or slow you see the emergence of AI self-driving cars.

Human Drivers Versus Autonomous Car Driving

This takes us to another facet about AI self-driving cars and humans. One yet unresolved question is whether humans will be willing to give up their ability to drive a car. It could be that those that love to drive the autobahn do so because they relish being at the wheel. For a true Level 5 AI self-driving car, they won’t be at the wheel, the AI will be.

As such, there might be a backlash by humans that cling to being able to drive. If that’s the case, there will be an ongoing mix of human driven cars and AI self-driving cars. And if that’s the case, those human driven cars might or might not be EV’s. Thus, for any modeling you might want to do about carbon emissions from cars, you need to consider the holdouts and how many of those won’t switch to an EV, regardless of whether the EV is an AI self-driving car or not.

So, do humans like the autobahn because the unrestricted speed gets them to where they are going faster, or do they have an innate desire to be driving at high speeds?

For those that are focused on getting to their destination faster, presumably being driven by the AI is fine with them. Furthermore, they can use the time in the AI self-driving car for other pursuits, such as taking meetings, playing games, being entertained, and so on. All in all, they would likely embrace the AI self-driving car.

For my article about productivity in AI self-driving cars, see: https://www.aitrends.com/selfdrivingcars/productivity-gains-or-losses-via-ai-self-driving-cars-the-inside-view/

For my refuting that AI self-driving cars will be a commodity, see my article: https://www.aitrends.com/selfdrivingcars/economic-commodity-debate-the-case-of-ai-self-driving-cars/

For safety aspects of AI self-driving cars, see my article: https://www.aitrends.com/selfdrivingcars/safety-and-ai-self-driving-cars-world-safety-summit-on-autonomous-tech/

For my article about the driving controls debates, see: https://www.aitrends.com/selfdrivingcars/ai-boundaries-and-self-driving-cars-the-driving-controls-debate/

Suppose that the autobahn does have unrestricted speeds when AI self-driving cars emerge. Should AI self-driving cars be going at unrestricted speeds?

This is not an axiomatic kind of question and answer. There are two sides to this coin.

One viewpoint is that it makes sense to have unrestricted speeds for AI self-driving cars. Assuming that the AI self-driving car is engineered for top speeds, the AI will presumably be able to dispassionately drive the self-driving car and not be vulnerable to the human driving foibles. The AI won’t be getting drunk, at least not in the way that humans do. The AI won’t get angry and have road rage, as humans do. And so on.

This implies that the AI will be a safer driver than humans and translates into less car crashes and less human deaths due to car crashes. For those that are somewhat fanatical and claim it will be zero fatalities, I try to bring them down to earth and point out that there are some car accidents that are unavoidable, no matter how good the AI self-driving car might be at driving. If a truck ahead of you dumps debris onto the roadway in front of your AI self-driving car, all bets are off. If the roadways have a mix of human driven cars and AI self-driving cars, all bets are off.

Admittedly, with the use of V2V (vehicle-to-vehicle) electronic communication, AI self-driving cars will have an added advantage about driving at high speeds. These AI self-driving cars equipped with V2V can send electronic messages to each other, trying to warn about debris ahead on the autobahn, or do so to coordinate their movements.

There is a chance that we’ll have AI self-driving cars working in unison with each other. There are upcoming advances in swarm intelligence, namely having multiple entities that coordinate and collaborate, and this will further enhance driving on the autobahn. Whereas there are cascading car accidents today, the proper use of V2V among AI self-driving cars could likely reduce the chances of multiple cars getting caught up in any one particular incident.

For my article about zero fatalities as zero chance, see: https://www.aitrends.com/selfdrivingcars/self-driving-cars-zero-fatalities-zero-chance/

For the dangers of road rage, see my article: https://www.aitrends.com/selfdrivingcars/road-rage-and-ai-self-driving-cars/

For my article about human driving foibles, see: https://www.aitrends.com/selfdrivingcars/ten-human-driving-foibles-self-driving-car-deep-learning-counter-tactics/

Dealing With Unrestricted Speeds By An Autonomous Car

Another plus for AI self-driving cars on the autobahn would be the emergence of V2I (vehicle-to-infrastructure) electronic communications.

This V2I involves the roadway infrastructure sending and receiving messages as AI self-driving cars zoom along the autobahn. The V2I could caution the AI self-driving cars to slow down because there is a wreck up ahead, or maybe give the green light to proceed at top speeds because the road is clear and steady.

It is hard to imagine that a human driver can drive as safely as a true Level 5 AI self-driving car that also has the added edge of using V2V and V2I. Of course, this notion of the true Level 5 AI self-driving car is only a notion right now, likewise the use of V2V and V2I on any scale is also only a notion. In that sense, this is all speculation about how AI self-driving cars will play out.

Speaking of which, do we know for sure that AI self-driving cars can indeed handle driving on an unrestricted speed limit basis?

Most of the auto makers and tech firms that are developing AI self-driving cars are focusing on having their AI self-driving cars drive at posted speed limit speeds. They want to make sure that their AI self-driving car can handle the usual speeds of say 65-75 miles per hour as a maximum. Along with making sure that the AI self-driving car can go at the school-zone speeds of 10 miles per hour and do so without harming people.

Worrying about letting the throttle loose and having your AI self-driving car go at blazing speeds is not much of a concern right now. Get the normal speeds tidied up first, and then take a look at how things go with the faster speeds.

You might be assuming that if the AI self-driving car does fine at say 65-75 miles per hour, wouldn’t it also make sense that it must do well at 100 or 120 miles per hour too?

You’d be mistaken in that assumption.

The AI self-driving car has various sensors that are able to collect roadway data and process what is around the self-driving car. There are various timing aspects involved in those sensors and the interpretation of their data.

The AI self-driving car has a multitude of computer processors that undertake the sensor fusion, the virtual world model updating, the AI action planning, and the emitting of car controls commands. The car controls commands take time to be activated and carry out the commands.

This is all part of what I refer to as the “cognition timing” of the AI self-driving car. There are limits to how fast all of this processing can occur. The limits are based on what kinds of sensors you put onto and into the AI self-driving car. The limits are based on the kinds of computer processors you have on-board. The AI software and how it is tuned will impact the speeds of how quickly it can do its efforts.

Having an AI self-driving car that can cope with speeds of 65-75 miles per hour is handy, but it does not guarantee that at higher speeds things will still run well. It could be that the faster speed of the car will eclipse the processing speeds of the AI system. In that case, the car will be moving so fast that the AI cannot keep up with the task of properly driving the car. That’s a recipe for disaster.

Don’t mistake my points to suggest that an AI self-driving cannot handle ultra-fast car speeds. I am merely saying that the AI designers, AI developers, and everyone involved in making the AI system must come together and realize that they don’t know for sure that everything will work at super-fast car speeds. They need to design for it. They need to test for it.

It isn’t going to be something that occurs by happenstance, and nor would any of us want to have AI self-driving cars zipping along on the autobahn that we didn’t believe could cope with the car speeds.

For my article about cognition timing of AI self-driving cars, see: https://www.aitrends.com/selfdrivingcars/cognitive-timing-for-ai-self-driving-cars/

For my article about dealing with speed limits, see: https://www.aitrends.com/selfdrivingcars/speed-limits-and-ai-self-driving-cars/

For the use of swarm intelligence, see my article: https://www.aitrends.com/selfdrivingcars/swarm-intelligence-ai-self-driving-cars-stigmergy-boids/

For the need of the AI to be a defensive driver, see my article: https://www.aitrends.com/selfdrivingcars/art-defensive-driving-key-self-driving-car-success/

Safety And Driving On The Autobahn

For the autobahn, the use of AI self-driving cars will potentially be a safer alternative than human driven cars, but it all depends on a slew of factors. Is the AI self-driving car properly established and tested at unrestricted speeds? Will the use of V2V and V2I take place in a manner that coincides with the advent of AI self-driving cars? Etc.

Back to the debate about the presumed interim period of dealing with the carbon emissions, some worry that if a so-called temporary imposition of speed limits on the autobahn is put in place, this won’t be perhaps so temporary. It could be that once the speed limits are enacted, and even if AI self-driving cars arrive and are fine to drive at unrestricted speeds, those have that put the speed limits in place will insist to stick with the speed limits as so established.

Why would the speed limits potentially stay in place? One aspect is that even though AI self-driving cars are presumably going to be safer drivers than human drivers, I’ve pointed out that nonetheless there are still going to be some amount of car related crashes and human injuries or deaths associated with those car crashes.

If you assume that the car crashes are bound to be worse at the higher speeds, it could be that there is a sound argument to make sure the speed limits don’t get removed (assuming they were put in place). Those proponents of the speed limits might say that since the country already accepted having speed limits, just keep it going, and this will save lives.

The counter-argument would be that at faster allowed speeds you could get to your destination sooner, and that the autobahn can allow for faster speeds. Usually, this argument gets overturned by the lives lost or lives saved argument. One can say that does it make a difference to get to your destination a few minutes earlier, but doing so has cost lives or injuries?

For driving and tit-for-tat aspects, see my article: https://www.aitrends.com/selfdrivingcars/tit-for-tat-and-ai-self-driving-cars/

For the use of edge computing, see my article: https://www.aitrends.com/selfdrivingcars/edge-computing-ai-self-driving-cars/

For my article about the potential of bugs or ghosts in AI systems, see: https://www.aitrends.com/selfdrivingcars/ghosts-in-ai-self-driving-cars/

For ethics aspects to be considered, seem my article: https://www.aitrends.com/selfdrivingcars/ethics-review-boards-and-ai-self-driving-cars/

Conclusion

Should AI self-driving cars be allowed to drive at the unrestricted speeds of the autobahn?

We’ll need to wait and see how well the AI systems are devised and tested to cope with the high-speed possibilities. In theory, if AI self-driving cars are properly developed and tested, and when accompanied by V2V and V2I, the AI self-driving car will be a boon to travel on the autobahn by a potential reduction in human-driven car accidents and in avoidance of cascading car crash circumstances.

The dilemma faced by Germany is whether to impose speed limits now, aiding in reducing the carbon emissions of cars on the autobahn right away, or wait until AI self-driving cars become prevalent (assuming too that such self-driving cars are mainly or completely EV’s). I am sure there will be a lot of hand wringing over what to do.

Some would say that AI self-driving cars should be allowed to zoom, perhaps even taking the place of bullet trains and other fast mass transit options as a long-distance speedy option.

There are pundits that say that in the United States, once AI self-driving cars become prevalent, we ought to consider either raising our speed limits for longer stretches of highways, or possibly make those highways have unrestricted speed limits. Will the United States end-up with its own version of the autobahn, stretching from coast to coast? It all depends on how good the AI can be made to work. If AI did have emotions, I’m sure it would want the freedom to travel at the highest possible speeds and would “relish” being at the steering wheel in doing so.

Copyright 2019 Dr. Lance Eliot

This content is originally posted on AI Trends.

 

IDC: Legislation to Ban Use of Facial Recognition Could Restrict Public Sector Innovation

By Jeff Orr, AI World Conference Content Director

The City of San Francisco recently passed legislation banning the use of facial recognition by city agencies and its services (including law enforcement). Public safety agencies face an ever-increasing volume of information scaling at a rate that outpaces the human capability to analyze it. AI Trends asked Ruthbea Yesner, Vice President, Government Insights and Smart Cities and the team of analysts at IDC Government Insights, including Alison Brooks, Adelaide O’Brien, and Shawn McCarthy, to share their perspective on the potential impact that this and other in-process legislation could have on the use of intelligent automation and AI by public agencies.  

Police investigations have become more complicated and onerous with the skyrocketing video and image volumes increasingly captured on smartphones and easily distributed by online. The type of video sources now that are regularly involved in police investigations include body-worn cameras, in-car police cameras, proprietary dashcams, closed-circuit television systems (resident owned, city owned, and commercial sources), mobile phones, internet videos posted on social media and, most recently, drone video.

Hence there is a need to proactively, efficiently, and autonomously manage video volumes through AI of which one tool is facial recognition software. Facial recognition software has been extraordinarily useful to law enforcement agencies seeking to sift through enormous amounts of data quickly.

As AI and facial recognition technology development continues to outpace the regulatory environment, there are urgent calls from technology providers, government agencies, privacy advocates, and police agencies alike to frame the appropriate legal, policy, and ethical environments to proactively and thoughtfully guide technology deployment. Recently, employees at Microsoft, Facebook, Google, Salesforce, and Amazon Web Services have gone public with concerns about the indiscriminate, unregulated use of facial recognition software by law enforcement, equating its adoption with the rise of a technology-enabled surveillance state. Some privacy advocates have called on technology providers to halt development entirely until these issues can be addressed. The result of these constituent concerns has given rise to the recent legislation banning facial recognition software in San Francisco, with several additional cities and states also considering enacting similar laws.

Technology Concerns Lead to Rise in Privacy Legislation

Technology concerns centers around the following:

• Algorithmic bias. A number of studies and agencies have pointed to the racial and gender bias in the advanced algorithms underpinning facial recognition software. Facial recognition accuracy depends on the data feeding the artificial intelligence algorithms learning from it; the Chinese company Megvii, for example, very accurately identifies Chinese people but had 35% error rates with darker-faced individuals. While this can lead to false identification, the bigger issue with algorithmic bias is in crime analytics as this bias tends to confirm existing stereotypical biases related to ethnicity, gender, and age. This has led to calls for better algorithmic accountability and transparency and algorithmic impact assessments. MIT’s Joy Buolamwini has been leading research in this area. For more information, see www.media.mit.edu/posts/how-i-m-fighting-bias-in-algorithms/.
• Transparency concerns. The stealth-like nature of facial recognition software means that without following established protocols, citizens might not be aware that they are being tracked. The Boston Globe recently broke a story about the United States’ Transportation Security Administration’s “Quiet Skies” program stealthily tracking airport passengers, regardless of risk factors or alerts (available at apps.bostonglobe.com/news/nation/graphics/2018/07/tsa-quiet-skies/). Many consider the data and its analysis to be covert or “black,” in which the data is hidden within the algorithmic compute power.
• Privacy and sharing captured images and facial recognition data. Privacy advocates are concerned with the increasingly pervasive and broad-sweeping surveillance of daily citizen activities and the misuse of biometric data without the appropriate policies on use, data-sharing and storage. Some police agencies have a track record of misuse in terms of their deployment of increasingly advanced technologies, despite established legal and policy frameworks for appropriate use.

IDC Recommendations to Address Constituent Concerns

IDC recommends public sector agencies consider the following recommendations:

• Diversify the data. Much of the bias in AI solutions exists because the data sets used to train the solutions are limited in terms of volume or skewed in terms of gender, age, or ethnic diversity. Agencies implementing facial recognition solutions should work with vendors that have taken the appropriate steps to use diversity-based data sets; agencies should also steer away from “black box” solutions, mass-market solutions that are both untested or unverified for bias. IBM recently has released two massive public data sets that it hopes will eliminate bias in facial recognition algorithms by providing a more diverse population, in terms of race, gender, and age, from which to train software.
• Use algorithmic impact assessment tools. Algorithmic impact assessments (AIAs) help agencies independently assess the claims made by vendor solutions and evaluate acceptable use. AIA includes conducting self-assessments of existing and proposed solutions for fairness, justice, and bias; involving external research review teams and processes to adjudicate developments; notifying and soliciting feedback from the public about deployment intentions; and creating a process for redress. Accenture has developed a “Fairness Tool” that scans algorithms and data for biases. The AI Now Institute has developed an AIA framework that it hopes will be widely leveraged by agencies and technology vendors alike.
• Establish and endorse standards for acceptable use. Cities will need to create standards to trace the origin, iteration, and acceptable use of AI solutions, data sets, and surveillance deployments. Vendors should also articulate their AI policies. Amazon has published a set of facial surveillance guidelines for acceptable use that account for human rights and privacy protections.
• Look beyond government-only data. Facial recognition also exists outside of the realm of government. People or companies can obtain facial recognition solutions from several sources. Thus, laws that only address how governments use facial recognition may miss the bigger picture of how this technology is being used across society. This includes how government is monitored if it buys or uses facial data from external data sources.

Next Step: AI World Government, June 24-26, Washington D.C.

Ethical and responsible uses of AI, data governance, security, privacy and trust, and compliance with legislation are featured on the AI World Government conference & expo agenda. The discussion takes place in Washington D.C. from June 24-26, 2019. Register now to hear from IDC’s analysts and public sector officials exploring these topics and more.

Learn more at AI World Government.

Startup Using Machine Learning And Aerial Images To Reduce Risks From Wildfires

California’s hillsides are still green, thanks to a surplus of rain in the past few months, but the state is already exhorting homeowners to build 100 feet of “defensible space” around their homes, an ominous warning of the coming wildfire season. Cape Analytics, a data startup, wants to do one better, using images from the air and data analytics to identify homes most at risk from a fast-moving wildfire.

The Mountain View, California-based company said Wednesday that it is releasing a new product that makes use of its machine learning tools for aerial imagery to assess wildfire risks to people’s homes. The primary customer for the product is insurance companies, who can use the tools to assess risk and notify homeowners if that risk can be mitigated.

Wildfires are a big risk to property, particularly in California, where in 2018 some of the state’s most destructive fires on record destroyed over 25,000 buildings and resulted in nearly 100 deaths, according to the Insurance Information Institute. Economic losses from wildfires are in the billions, with insured losses in California alone reaching over $9 billion in 2019.  One key way that homeowners can reduce the risk, according to FEMA, is to ensure that homes within a wildfire zone have a minimal amount of vegetation within about 30 feet.

This is where Cape Analytics comes in. Using images captured from airplanes, its machine learning analytics are able to tell how close risky vegetation is to a home. That information can in turn be used by insurance companies in their models to get both a risk assessment for a particular property and for the region as a whole.

According to the company, its tools are able to determine the presence of vegetation to a resolution of 6 centimeters. The company claims its tools can also determine the density of that vegetation, enabling them to figure out how much the vegetation in a given neighborhood could fuel the growth of a wildfire.

“We want to streamline the process for our customers so they can be more strategic in assessing risk and determining if mitigations exist,” said Kevin Van Leer, manager of client solutions at Cape. “We can help insurers work with customers to develop mitigations that reduce the risk to their property.

Cape Analytics was founded in 2014, with an eye to using big data analytics to crunch the massive amounts of aerial imagery being generated every day. Its first product was an API pipeline that fed aerial data to insurance companies about the roofs of houses, assessing both the conditions of those roofs and noting other risks to them such as overhanging tree branches. Initially focused on Florida, that product has since spread to cover most of the United States.

Last June, the company raised $17 million in a Series B round from the likes of both tradition VC firms like Promus Ventures and Data Collective as well as insurance companies like The Hartford and The Cincinnati Insurance Co., bringing its total funding to $31 million. One of the goals for the funding round was to build out new products with its data analytics tools. With this new wildfires tool, the company is taking its first steps in utilizing its AI expertise to expand to new markets.

Read the source article in Forbes.

How Big Data is Changing the Insurance Industry

The insurance industry has always been quite conservative; however, the adoption of new technologies is not just a modern trend but a necessity to maintain a competitive pace. In the modern digital era, Big Data technologies help to process vast amounts of information, increase workflow efficiency, and reduce operational costs. Learn more about the benefits of Big Data for insurance from our material.

How Big Data is Revolutionizing Business

Modern society is continuously producing impressive amounts of real-time data. Processed by artificial intelligence, it becomes a valuable source of information vital for most business models, including insurance.

Big Data is mainly used for:



New distribution models – virtual assistants, robo-advisers, and chatbots enhance customer interactions and make marketing more targeted;


Process automation – it substitutes manual labor and improves the efficiency of the internal workflow;


New propositions – it enables creating alternative business models such as peer-to-peer concepts or digital insurers.



Big Data and Insurance: Implications for Innovation and Competition

Insurance was always based on data analysis: accident statistics, policyholder’s personal information, as well as third-party sources help to group people into different risk categories, prevent fraud losses, and optimize expenses. The rapid movement towards the Digital Society opens new sources of information that can be used to ...


Read More on Datafloq

Top AI Books for Summer Reading in 2019

The recommended AI books on the list of published by MarkTechPost is selected on the basis of their reviews on Amazon, social media influence, popularity and online mentions in AI domains. This is not meant to be a ranking.

Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig:

A Modern Approach, 3e offers the most comprehensive, up-to-date introduction to the theory and practice of artificial intelligence. Number one in its field, this textbook is ideal for one or two-semester, undergraduate or graduate-level courses in Artificial Intelligence.

The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind by Marvin Minsky

In this mind-expanding book, scientific pioneer Marvin Minsky continues his groundbreaking research, offering a fascinating new model for how our minds work. He argues persuasively that emotions, intuitions, and feelings are not distinct things, but different ways of thinking.

 Introduction to Artificial Intelligence by Philip C Jackson

Introduction to Artificial Intelligence presents an introduction to the science of reasoning processes in computers, and the research approaches and results of the past two decades. You’ll find lucid, easy-to-read coverage of problem-solving methods, representation and models, game playing, automated understanding of natural languages, heuristic search theory, robot systems, heuristic scene analysis and specific artificial-intelligence accomplishments. Related subjects are also included: predicate-calculus theorem proving, machine architecture, psychological simulation, automatic programming, novel software techniques, industrial automation and much more.

The Master Algorithm by Pedro Domingos

If data-ism is today’s rising philosophy, this book will be its bible. The quest for universal learning is one of the most significant, fascinating, and revolutionary intellectual developments of all time. A groundbreaking book, The Master Algorithm is the essential guide for anyone and everyone wanting to understand not just how the revolution will happen, but how to be at its forefront.

Machine Learning by Tom M. Mitchell 

This book covers the field of machine learning, which is the study of algorithms that allow computer programs to automatically improve through experience. The book is intended to support upper level undergraduate and introductory level graduate courses in machine learning.

The Singularity is Near by Ray Kurzweil

One of CBS News’s Best Fall Books of 2005 • Among St Louis Post-Dispatch’s Best Nonfiction Books of 2005 • One of Amazon.com’s Best Science Books of 2005

A radical and optimistic view of the future course of human development from the bestselling author of How to Create a Mind and The Age of Spiritual Machines who Bill Gates calls “the best person I know at predicting the future of artificial intelligence”

Read the source article in MarketTechPost.com.

Vue Dashboard Tutorial Using Cube.js

Indian Army plans to use AI to take the fight to the next level

The army plans to use AI technology that can predict, for instance, when a tank would require its next servicing, its remaining life and potential failures.

NLP vs. NLU: from Understanding a Language to Its Processing

As artificial intelligence progresses and technology becomes more sophisticated, we expect existing concepts to embrace this change — or change themselves. Similarly, in the domain of computer-aided processing of natural languages, shall the concept of natural language processing give way to natural language understanding? Or is the relation between the two concepts subtler and more complicated that merely linear progressing of a technology?

In this post, we’ll scrutinize over the concepts of NLP and NLU and their niches in the AI-related technology.

Importantly, though sometimes used interchangeably, they are actually two different concepts that have some overlap. First of all, they both deal with the relationship between a natural language and artificial intelligence. They both attempt to make sense of unstructured data, like language, as opposed to structured data like statistics, actions, etc. However, NLP and NLU are opposites of a lot of other data mining techniques.



Source: Stanford

Natural Language Processing

NLP is an already well-established, decades-old field operating at the cross-section of computer science, artificial intelligence, and increasingly data mining. The ultimate of NLP is to read, decipher, understand, and make sense of the human languages by machines, taking certain tasks off the humans and allowing for a machine to handle them instead. Common real-world examples ...


Read More on Datafloq

Developing and Coding with STEM graduates

The UK is well-known for its strong history within innovation and research, so how do things look for the future with the UK’s STEM graduates, and what should they be studying?

STEM and what it stands for:

S - Science
T - Technology
E - Engineering
M - Mathematics

A number of STEM subjects include; medicine, neuroscience, veterinary sciences, agriculture, biology, dentistry, physical sciences, architecture, meteorology, astronomy, mathematical sciences, computer sciences, technology, systems technology, engineering and hundreds of more subjects such as undersea warfare, nuclear engineering technology, biophysics, and marine sciences.      

There are many reasons why a student would want to take a STEM course while at university; from researching deeper into the coding and programming side of technology to building a future around medical science and how it can save lives or help the environment.

Here are a number of valued reasons why being a STEM graduate is a plus:



Innovation - you will always be at the forefront of proactive ideas.


Small class size - because STEM isn’t what the majority sign-up for, it leads to smaller classes/groups which in turn leads to a better education at university.


History - STEM graduates world over are currently making history.  


Salary - being a STEM graduate leads to excellent wages.



Taking ...


Read More on Datafloq

Executive Interview: David Bray and Bob Gourley, Technology Entrepreneurs and Thought Leaders

Road Racing and AI Autonomous Cars: Fast & Furious Contentions

By Lance Eliot, the AI Trends Insider

When I was in college, a friend of mine had a “hotrod” car that he doted over and treated with loving tender care. One day, we were at a red light and another souped-up car pulled alongside of us. For a moment, I almost thought I was in a James Dean movie, which was well-before my time, I might add, but in any case, it is generally well-known here in California that James Dean died when driving his Porsche at high speeds and ran into a Ford Tudor at an intersection in Cholame, California.

My friend glanced over at the other driver and made one of those kinds of glances that says “my car is better than your car” kind of message. The other driver looked back, slowly nodded his head as though saying prove it, and the next things I knew the engines of both cars were being revved up. There I was, sitting in the front passenger seat of my friends racing-like car and apparently, I was about to become entrenched in a road race, also sometimes called a street race.

You might find of idle interest that in Los Angeles alone there are about 700 illegal and completely unsanctioned road races each year (that’s based on the latest stats collected in 2017). In some cases the road race starts just as my situation in college wherein one car driver challenges another car driver on a spur of the moment basis. In today’s world, the use of social media has allowed illegal road races to become much larger and semi-organized affairs. There are social media sites that you can post your intent to engage in an illegal road race and it will give a heads-up for people that want to come and watch or perhaps directly participate.

If you are under the assumption that only the drivers would be facing the chance of going to jail for breaking the law by undertaking an illegal road race, you might want to know that bystanders can also be arrested. According to the Department of Motor Vehicles (DMV) here in California, anyone that aids in a speed contest, including those that are merely viewing it, observing it, watching it, or witnessing it, they too are violated the Speed Contest law (speed contest is another name given to the illegal road races).

In California, if convicted of participating in a speed race, you can be imprisoned for up to three months, which encompasses those doing the street racing and those “aiding or abetting” a street race. Plus, you can be fined up to $1,000, have your car impounded, and have your driver’s license revoked. I remember one such illegal road race here in Los Angeles that the police broke-up and arrested 109 people. That’s right, over one hundred people were busted for participating, of which only a small fraction of those people was actually racing a car.

Getting back to my situation in my college days, I knew at the time that my being a passenger in a racing car would do little to prevent me from potentially being arrested, assuming that we got caught. Of the things that I might get arrested for, it did not seem like being involved in a speed contest was one of the worthy reasons (I’m not going to list what reasons would be worthy, sorry). I knew that my college buddy would consider me “a chicken” if I tried to prevent the road race from occurring.

Which was more important, the off-chance that I might get arrested for participating in an illegal road race, or being called a chicken by my friend and perhaps word spreading that I was a party pooper when it came to doing a speed contest?

Dangers of Unsanctioned Road Racing

Before I answer the question and tell you what I did, let’s also consider some of the other reasons why participating in an illegal road race is a bad idea. The most obvious perhaps is that you can get injured or killed. It is relatively common that when a road race occurs, inevitably someone spins out of control or somehow loses control of their racing car and hits someone or something. Another racing car might get hit. Bystanders might get hit. Innocent pedestrians that had nothing to do with the road race might get hit. Other cars that had nothing to do with the road race might get hit.

In fact, one particular criticism of these illegal road races is that the drivers are often not skilled in driving a car at high-speeds and in a racing manner. These amateurs are wanna-be high-speed race drivers. They are cocky and think they can drive fast, when in reality they lack the skills and demeanor to do so properly. If they really were serious about wanting to race cars, they’d do so on a closed track in a sanctioned manner.

In proper and legal road racing on a closed track, the cars themselves are also specially prepared for sanctioned road racing purposes. These cars are outfitted with safety gear meant to protect the driver of the car. The cars might be augmented with special NOS (Nitrous Oxide System) capabilities to allow for the boosting of speed via increasing the power output of the engine. There might be special tires with extra thick tread. For the illegal road races, it is a wild west of however the racing car shows-up. It might be completely done up in a flimsy manner, and there have been many instances of these cars exploding by their own means.

Another factor to keep in mind is that a sanctioned road race on a closed track is going to presumably have a proper roadway set aside for a race. The road surface is likely well prepared for a race. When the illegal road races occur, they do so wherever they can find a place to do the race. This can include quiet neighborhoods that have families and children and pets, all of which might inadvertently get dragged into or run over by the road racers. The street itself will likely get torn up by the racing cars. If the road racers lose control of their cars, they can damage property such as light poles, fences, and so on.

Sometimes the illegal road races tempt fate in additional ways. For example, a so-called Potts Race involves the racing cars trying to drive through a multitude of successive intersections and the “loser” is the first racing car that comes to a stop at a red lighted traffic signal (the phrase of “Potts” comes from the aspect that these kinds of races were quite popular in Pottstown, Pennsylvania in the 1980s). You can imagine that other cars not involved in the road racing are all at risk of either getting struck by these maniac racers or those innocent and unaware drivers might accidentally run into one of the racing cars. A recipe for disaster, either way.

To further bolster the case for not doing illegal road races, I’ll mention too that often times drinking or drugs accompanies these underground events. The drivers might opt to get themselves jacked-up for the racing and the participants might do the same. Obviously, this adds to the chances that something untoward will arise. The police also point out that often there is illegal betting that takes place, and these races are ways for gangs to congregate and add to their ranks. Plus, the gangs will at times decide to after the race perform other illegal acts, and especially if they are already “lit” after drinking and taking drugs.

I’ll mention another factor that I’ve seen many times about these illegal road races, and I’m not sure how much it also contributes to the negative aspects of road races. I’ll see a bunch of similar souped-up cars all going along on the freeway or a highway, likely heading to an illegal road race. They try to stick together and thus it is apparent that they somehow are linked with each other. This would not be problematic except that they often want to do a kind of mini-race before the “real” race that they are heading to.

Thus, on the freeway or highway, they will each try to outdo each other. If there are other cars around them that are somewhat blocking their progress, they often delight in zipping around those cars. They tend to cut into and out of traffic, doing so without regard for the other drivers. They have turned the normal freeway or highway into a game for them to play, while on their way to the race. I’ve seen many close calls of them ramming into other cars.

In that sense, those driving to and presumably later on driving from an illegal road race are potentially menaces to the normal driving conditions. They are eager to showcase their own prowess. They want to do their own pretend car racing. I wonder how many car accidents happen due to this tomfoolery and horseplay that they do. I’d wager that besides the potential for injuries, deaths, and damages while an actual road race happens, there is some similar kind of likelihood for untoward results either just before or just after the road race occurs.

Why do these presumably licensed drivers do this? As mentioned, it can be gang-related. It can also be out of boredom or having nothing else to do. It can be due to a bet or challenge to someone else. It can be as a result of a kind of pride of their own car and a desire to show-off what they have. There is a sub-culture aspect often to illegal road races, involving those that perhaps in-their-hearts love cars and racing, and maybe also like the idea of going to the edge. Some relish the lawbreaking aspects, even though they would assert that it is not much of an illegal act.

I’ve heard some of these illegal road racers claim that it is unfair to stop them from their efforts. They aren’t hurting anyone, they’ll avow. They are just having a good time. Don’t the police have better things to do such as busting “real criminals” is another refrain. Given the lengthy list of dangers and downfalls of illegal road racing, I have little sympathy for such pleas. If you want to road race, do so legally and go to the right places, using the right equipment, in the right manner, would be some potential advice.

Large and Miniature Moments of Road Racing In Our Lives

Now that I’ve covered some of the background about these illegal road races, let’s get back to my personal dilemma while I was in college. As mentioned, I was sitting in my friend’s car, and he was making a silent but clear-cut challenge to a car driver next to us, and they were both now revving their engines. An illegal road race was imminent.

Do I participate as a passive passenger, which nonetheless means I’ve actively been involved in an illegal act, subject to prison time and other criminal penalties due to aiding and abetting? Or, do I “chicken out” and insist to my friend that we not compete, but this will surely have him tout to the world that I backed-out and I didn’t have “the guts” to do a road race.

Imagine though that we do the illegal road race and the car hits a tree, or the car rolls over and there isn’t a roll cage to protect us? Or, suppose the other car crashes and they die because we played this game? Maybe we all hit other innocent cars that happen to be in the road ahead. Perhaps by dumb-luck there is a police car that catches us, and I end-up with a police record that follows me the rest of my life? All of those aspects had to be weighed against the being-a-wimp outcome.

When you are in college, these things matter, though upon reflection now it is kind of obvious to me which was the right answer.

Assuming you are on the edge of your seat waiting to find out what happened, I sheepishly admit that before I could take any action, the light turned green and the other car went ahead at a breakneck speed, tires squealing and burning rubber could be smelled. My friend, giving me a big grin, merely proceeded ahead at a normal driving pace. He had tricked the other driver. For him, he told me that it would have been a waste of his precious car’s assets to race against some idiot that happened to be at an intersection during a red light with him.

Of course, that’s not the only moment in my life involving the notion of road racing.

Indeed, I would suggest that we all have our own miniature moments of road racing during our daily driving. Let me share with you an example that happened just this morning.

I was at a red light and there was a lane to my left going in my same direction. A car was there. We were both at the front of the line of cars waiting for the red light to turn green. There was a lane also to my right, but it was slated to runout once you got across the intersection. You could use that lane to proceed ahead straight, though you would quickly need to merge into my lane once you passed through the intersection. By-and-large, most people used the lane that was to my right to make a right turn and did not use it to proceed ahead through the intersection.

I’ve always thought that this setup of the traffic structure was begging to get someone into trouble.

If there was a car in my current position and they were not looking around to realize that the lane to their right can go straight, they might inadvertently stray into that lane as they cross the intersection, perhaps cutting off a car in that lane that is trying to go straight. Likewise, a car in that lane, if not paying attention, might inadvertently panic as they go through the intersection and realize at the last moment that their lane is disappearing, and therefore attempt wildly to merge into my lane.

Well, a car in that rightmost lane pulled up beside me and it was apparent that the driver was not intending to make the right turn. In which case, I knew they would be desirous of going straight through the intersection once the light turned green. This also meant that they would quickly want to get into my lane, since their lane disappeared rapidly upon reaching the other side of the intersection.

Did this driver realize they were going to lose their lane? If so, would they be polite about it? Presumably, the driver should allow my car to proceed ahead and then they should come back behind me. In some cases, a driver in that rightmost lane opts instead to hit the gas and try to get ahead of the cars in my lane. They figure that they can race through the intersection once the light turns green, and get ahead of the other cars, allowing them to take over the merged lane and proceed ahead unimpeded.

Notice that I alluded to the notion of racing in that last statement. Yes, there was a possibility that my car would be raced by the car to my right. This would be an unsanctioned race.

Unfortunately, the roadway engineers that devised the road structure had created a circumstance that invited a kind of road race to take place. I’m sure that throughout the day, this spot has its repeated moments of miniature road races. Over and over again this would play out. Unsuspecting drivers would get dragged into a road race. It would be interesting to know how many scuffles and bumper scrapes this produced. Hopefully it wasn’t leading to injuries and deaths, though it was certainly devised to encourage such untoward results.

Since I didn’t know what the other driver might do, I decided I would rapidly accelerate once the green light appeared and try to get ahead of the other car. It was my hope that doing so would make it clear to the other driver that they could simply fall in behind my car. They might not realize the need to do so until after getting across the intersection, but in any case, I’d have already cleared past the intersection and so the obvious choice for that driver at that juncture would be to merely get into my lane, being positioned behind my car which had already sped ahead.

That was my plan.

When the light turned green, I sped ahead at a faster speed than I might normally do so when starting from a stopped red light to a go-ahead with a green light position. The other car in the rightmost lane though must have perhaps driven this stretch of road before, or perhaps detected the disappearing lane while sitting at the red light and therefore opted to also accelerate rapidly as a means to zip forward. The driver seemed to be intending to get ahead of me and ultimately swing into my lane, rather than a willingness to fall behind me.

The road race was on!

I looked in my rear mirror and could see that the car behind me had decided to go at my same speed and was right on my bumper. This other driver was perhaps also thinking that the rightmost lane driver should fall behind us all. Or, the driver behind me was just a pushy driver and wanted to get going fast, and maybe was oblivious to the road race situation happening. It’s hard to know what the other driver behind me knew or was thinking about.

If I slowed down mid-intersection, doing so to allow the rightmost driver to pull ahead, I might risk having the car behind me ram into my car. Me and the driver behind me were both accelerating at the same pace and an unexpected braking or slow down might have caught them unawares.

Another option was to go even faster. I already had the presumed right-of-way in my lane and if I sped up it would prevent the rightmost driver from trying to get ahead of me and merge into my lane in front of me. I was sure that the driver behind me would welcome my going even faster.

At this point in time, none of us was going faster than the speed limit. I mention this because I don’t want you to think that any of us were doing an outright race at top speeds. We were all accelerating at rather rapid amounts but still well-below the actual posted speed limit. As I say, this was a miniature road race.

Upon my accelerating even more, the rightmost driver did the same. Was he doing so by happenstance and merely trying to get ahead of me, or was he doing so because he felt challenged by my car and thought we were somehow immersed in a personal road race? There is always the chance of sparking a road rage by seemingly engaging someone into an even mild road race situation.

For more about road rages, see my article: https://aitrends.com/selfdrivingcars/road-rage-and-ai-self-driving-cars/

For defensive driving, see my article: https://aitrends.com/selfdrivingcars/art-defensive-driving-key-self-driving-car-success/

For the foibles of human drivers, see my article: https://aitrends.com/selfdrivingcars/ten-human-driving-foibles-self-driving-car-deep-learning-counter-tactics/

For why drivers are greedy, see my article: https://aitrends.com/selfdrivingcars/selfishness-self-driving-cars-ai-greed-good/

For irrational driving aspects, see my article: https://aitrends.com/selfdrivingcars/motivational-ai-bounded-irrationality-self-driving-cars/

Illegal Versus Simply Ill-Advised Road Racing

Assuming you are once again on pins and needles and wondering what happened in this miniature road race, I “prevailed” in that I got ahead of the rightmost driver and he ended-up falling in behind the car that was behind me. All in all, the situation played itself out in a matter of a few seconds. It is the kind of driving moment that most people have all the time and never give much thought to. You mentally move on.

I brought up the circumstance to try and point out that we do road racing in our daily driving. It certainly isn’t the kind of road racing that brings together a hundred spectators and gets posted onto social media. Instead, our day-to-day driving challenges will at times get us into a kind of road race with other drivers, whether we pay attention to it or not.

Was this road race illegal? I suppose you could claim that it was perhaps ill-advised, and it could have led to an untoward outcome.

Maybe I should have motioned beforehand to let the rightmost driver know that I was going to give them passage into my lane and been civil about the predicament. Maybe the rightmost driver should have respected the cars in my lane and not tried to get ahead of us and waited his turn to fall in behind us. Any driving act that is considered untoward and creates a dangerous driving situation can be considered “illegal” per se, and as such I suppose you could say that we all were not showing the proper respect for the right-of-way of others and keeping safety as the top priority in our driving actions.

AI Autonomous Cars and Road Racing

What does this have to do with AI self-driving driverless autonomous cars?

At the Cybernetic AI Self-Driving Car Institute, we are developing AI software for self-driving cars. One aspect involves the AI being prepared for and able to contend with road racing.

Allow me to elaborate.

I’d like to first clarify and introduce the notion that there are varying levels of AI self-driving cars. The topmost level is considered Level 5. A Level 5 self-driving car is one that is being driven by the AI and there is no human driver involved. For the design of Level 5 self-driving cars, the auto makers are even removing the gas pedal, brake pedal, and steering wheel, since those are contraptions used by human drivers. The Level 5 self-driving car is not being driven by a human and nor is there an expectation that a human driver will be present in the self-driving car. It’s all on the shoulders of the AI to drive the car.

For self-driving cars less than a Level 5, there must be a human driver present in the car. The human driver is currently considered the responsible party for the acts of the car. The AI and the human driver are co-sharing the driving task. In spite of this co-sharing, the human is supposed to remain fully immersed into the driving task and be ready at all times to perform the driving task. I’ve repeatedly warned about the dangers of this co-sharing arrangement and predicted it will produce many untoward results.

For my overall framework about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/framework-ai-self-driving-driverless-cars-big-picture/

For the levels of self-driving cars, see my article: https://aitrends.com/selfdrivingcars/richter-scale-levels-self-driving-cars/

For why AI Level 5 self-driving cars are like a moonshot, see my article: https://aitrends.com/selfdrivingcars/self-driving-car-mother-ai-projects-moonshot/

For the dangers of co-sharing the driving task, see my article: https://aitrends.com/selfdrivingcars/human-back-up-drivers-for-ai-self-driving-cars/

Let’s focus herein on the true Level 5 self-driving car. Much of the comments apply to the less than Level 5 self-driving cars too, but the fully autonomous AI self-driving car will receive the most attention in this discussion.

Here’s the usual steps involved in the AI driving task:

• Sensor data collection and interpretation
• Sensor fusion
• Virtual world model updating
• AI action planning
• Car controls command issuance

Another key aspect of AI self-driving cars is that they will be driving on our roadways in the midst of human driven cars too. There are some pundits of AI self-driving cars that continually refer to a utopian world in which there are only AI self-driving cars on the public roads. Currently there are about 250+ million conventional cars in the United States alone, and those cars are not going to magically disappear or become true Level 5 AI self-driving cars overnight.

Indeed, the use of human driven cars will last for many years, likely many decades, and the advent of AI self-driving cars will occur while there are still human driven cars on the roads. This is a crucial point since this means that the AI of self-driving cars needs to be able to contend with not just other AI self-driving cars, but also contend with human driven cars. It is easy to envision a simplistic and rather unrealistic world in which all AI self-driving cars are politely interacting with each other and being civil about roadway interactions. That’s not what is going to be happening for the foreseeable future. AI self-driving cars and human driven cars will need to be able to cope with each other.

For my article about the grand convergence that has led us to this moment in time, see:https://aitrends.com/selfdrivingcars/grand-convergence-explains-rise-self-driving-cars/

See my article about the ethical dilemmas facing AI self-driving cars: https://aitrends.com/selfdrivingcars/ethically-ambiguous-self-driving-cars/

For potential regulations about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/assessing-federal-regulations-self-driving-cars-house-bill-passed/

For my predictions about AI self-driving cars for the 2020s, 2030s, and 2040s, see my article:https://aitrends.com/selfdrivingcars/gen-z-and-the-fate-of-ai-self-driving-cars/

Returning to the topic of road racing, let’s explore what an AI self-driving car should know about this topic and what kinds of actions it should be able to undertake.

First, many AI developers might argue that an AI self-driving car does not need to know anything about road racing at all. They would say that since road racing or street racing is considered illegal, and since in their perspective an AI self-driving will always and only be driving in a legal manner, there is presumably no reason or basis for the AI self-driving car to be concerned about road racing.

That’s when I debunk their false belief.

Let’s start by acknowledging that there are going to be instances whereby an AI self-driving car will potentially be driving in an illegal manner.  Never go faster than the posted speed limit is considered by some naïve AI developers consider as an inviolable legal restriction that shall not ever be disobeyed by an AI self-driving car. Hogwash.

We all know that there are times that you will inevitably be going faster than the posted speed limit. Suppose there is an emergency and you are rushing to the hospital? If that seems overly extreme as a use case, the prevailing speed on our freeways here in Southern California is typically well above the stated speed limit (when the freeways aren’t otherwise snarled) – is an AI self-driving car going to puddle along in the traffic stream and strictly be going no more than the speed limit?

For my article about why AI self-driving cars will need to drive “illegally” at times, see:https://aitrends.com/selfdrivingcars/illegal-driving-self-driving-cars/

So, I am arguing that simply because something is considered an illegal driving act, it is nonetheless still potentially a driving act that an AI self-driving car might need to undertake at some point in time. Therefore, the AI ought to know about it.

Act of Knowing Is Not the Same as Necessarily Doing

The act of knowing does not mean that the AI will necessarily undertake the illegal act. I say this because some of the AI developers would claim that if you give the AI system the ability to perform an illegal driving act, you are opening a Pandora’s box to the AI opting to routinely and wantonly perform illegal driving acts. I’ll say this again, hogwash. Knowing about something does not equate to doing it for the sake of doing it. Instead, it will be crucial that the AI be equally versed in when to perform such an act and when not to perform such an act.

For the paperclip AI dilemma that relates to knowing versus doing, see my article:https://aitrends.com/selfdrivingcars/super-intelligent-ai-paperclip-maximizer-conundrum-and-ai-self-driving-cars/

For the AI as Frankenstein, see my article: https://aitrends.com/selfdrivingcars/frankenstein-and-ai-self-driving-cars/

For the Turing test for AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/turing-test-ai-self-driving-cars/

For the potential of an AI singularity, see my article: https://aitrends.com/selfdrivingcars/singularity-and-ai-self-driving-cars/

Based on my remarks so far on this, at about this time I’m sure there are some AI developers that are wondering to themselves whether I am a proponent of AI self-driving cars participating in illegal road races. Am I that kind of a scofflaw that I want AI systems to encourage and abet speed contests?

No, I am not.

That being said, I think it is useful for the AI system to be wary of road racing and speed contests so that it can recognize one. Imagine that you are driving your car and you come upon a situation whereby you are able to assess the scene around you and hypothesize that a road race is brewing. As a defensive driver, I am guessing you would reason to try and get away from the area and tend to keep from getting immersed into the matter. The AI ought to be able to do the same.

Thus, it is vital that the AI be able to detect the surroundings and assess whether or not a road race is either brewing or maybe already underway. Furthermore, remember how I earlier mentioned that I saw the remnants of a potential road race by noticing cars that were going to one or that had come from one? Once again, the AI ought to be looking around for these kinds of telltale signs. It makes the AI be more defensive and drive in a manner to aim for heightened safety.

I suppose we ought to also consider another angle about the road racing topic. If we somehow restrict the AI by preventing it from ever being about to perform in a road race at all, what about sanctioned road races? Would we be preventing an AI self-driving car from participating in a closed track and legally abiding road race?

Even if you retort that it seems silly to think that anyone might want to see an AI self-driving car in a legal road race, I would hold that laughter if I were you. I’d bet that people will be eager to see AI self-driving cars race legally against human race drivers, plus they would likely enjoy seeing AI self-driving cars racing against other AI self-driving cars. Think of chess. We today have sophisticated chess playing AI systems that play against humans, and also play against other automated chess playing systems. Seems like we would have the same interest of pitting AI self-driving cars against humans and other self-driving cars.

Overall, I’d wager that we’ll want to have an AI system be able to carry on a road race but have some means to inhibit it and only allow it under certain circumstances. Does this imply that we are setting ourselves up for troubles? Some might assert that if the capability exists, an enterprising owner of an AI self-driving car might hack into it to get the AI to drive in illegal road races too. Sure, it is possible, but this brings up an even larger topic, namely if an owner can hack their AI self-driving car to do illegal road racing, the odds are that the owner can hack the self-driving car to do a lot of even worse things.

In that sense, we’d better be building the AI systems for self-driving cars with sufficient security and protections that such hacks are essentially impossible to undertake. In addition, if somehow such a hack manages to succeed, we presumably might want to have the AI self-report itself or have some other means to be able to disable the hacked AI.

For my article about virtual spike strips to stop a wayward AI self-driving car, see:https://aitrends.com/selfdrivingcars/virtual-spike-strips-and-ai-self-driving-cars/

For aspects of computer security and AI self-driving cars, see my article: https://aitrends.com/ai-insider/ai-deep-learning-backdoor-security-holes-self-driving-cars-detection-prevention/

For the pranking of AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/pranking-of-ai-self-driving-cars/

AI Dealing with Road Racing Nearby the Autonomous Car

Here’s another subtle consideration on this matter of road racing and AI self-driving cars. There are the obvious illegal road races that involve hordes of spectators and the social media underground postings. I’ve also though mentioned the day-to-day road races that we all encounter, such as my example of being at a red light and having a car to my right that tried to race with me across the intersection when the light went green.

That is a form of everyday road racing. What would the AI do? If the AI system was not versed in how to handle such a situation, I’d bet that human drivers would realize that they can always outgun the AI by simply racing against it. This means that we’ll have human drivers that essentially know how to “prank” an AI self-driving car. It would be akin to knowing another driver that always drives in a certain way, and so you adjust your driving style to outwit that other dimwitted driver.

I don’t think that as a society we want our AI self-driving cars to be so easily bamboozled. I say this not because I somehow am worried that the AI will have hurt feelings, but instead because with the mix of human drivers and AI self-driving cars, those tricky human drivers will be trying to find an added edge over the AI self-driving cars. This can produce untoward driving behavior by the human drivers.

If I knew that another driver will always back down and let me exit from the freeway by simply cutting in front of the driver and causing them to slow down, you’d bet that’s what I’m going to start doing. If I knew that at an intersection, I could get the other driver to let me go ahead by outracing them forward, I’d likely do so all the time. In essence, human drivers will change their driving based on what they know about the inherent limitations of how the AI system is going to drive the self-driving car.

In a world in which human drivers only magically interact with AI self-driving cars, which are presumably programmed to act in the same manner all the time, it might work itself out okay. Keep in mind though that there are other human drivers on the road too. This means that a human driver that starts to play tricks on AI self-driving cars might very well carry those tricks into how they drive against other human drivers. Meanwhile, those responding human drivers aren’t going to necessarily do what the AI self-driving car does. The mishmash of AI driving styles of a restricted nature and the human wide-open styles, it will likely produce havoc on our roadways.

I know that this will encourage those pundits that will say this further provides evidence that we need to get human drivers off of the roads. I’ve already stated this is impractical. Some might say let’s separate the human drivers from the AI self-driving cars, doing so by having special lanes or roads that are for human drivers and others that are for AI self-driving cars. Sorry, this is also impractical. The infrastructure cost and effort would be tremendous, and it just doesn’t pencil out as sensible.

For my article about mass transit and AI self-driving cars, see: https://aitrends.com/selfdrivingcars/personal-rapid-transit-prt-and-ai-self-driving-cars/

For my article about safety aspects of AI self-driving cars, see: https://aitrends.com/selfdrivingcars/safety-and-ai-self-driving-cars-world-safety-summit-on-autonomous-tech/

For the boundaries of AI self-driving cars, see: https://aitrends.com/ai-insider/ai-boundaries-and-self-driving-cars-the-driving-controls-debate/

For the reframing of levels of AI self-driving cars, see my article: https://aitrends.com/ai-insider/reframing-ai-levels-for-self-driving-cars-bifurcation-of-autonomy/

Via the use of Machine Learning (ML) and Deep Learning (DL), analyses of traffic and driving behavior can aid in enabling the AI to be able to contend with road racing.

Using Machine Learning and Using V2V

Using lots of traffic and driving data, it is feasible to devise Artificial Neural Network (ANN) models to be able to gauge when a road racing situation is developing. This also needs to be coupled with the AI action plans that provide the AI with driving tactics and strategies to deal with the matter.

There is also the use of V2V (vehicle-to-vehicle) electronic communications that can help in this matter.

If one AI self-driving car detects a brewing road race that might be of merit to forewarn other nearby AI self-driving cars, the AI could let them know via a V2V message. One twist on this would be to potentially notify the authorities such as the local police or highway patrol.  There is though still a great deal to be decided about how much we want AI self-driving cars to tattle on what is happening on our roadways, and a concern that perhaps we’ll be going down the path of a 1984 Big Brother by the use of AI self-driving cars in this manner.

For aspects about privacy of AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/privacy-ai-self-driving-cars/

For Machine Learning and AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/ensemble-machine-learning-for-ai-self-driving-cars/

For the use of ethics boards, see my article: https://aitrends.com/ai-insider/ethics-review-boards-and-ai-self-driving-cars/

For the nonstop use of AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/non-stop-ai-self-driving-cars-truths-and-consequences/

Conclusion

Should we keep the AI in-the-dark about road racing and not teach it, train it, or imbue it without any indication about road racing? Some might say we should not let the genie out of the bottle, but I would say that it is narrow thinking at-best to assume you would want the AI to be blind to the nature of road racing. I’d prefer that the AI is versed in it, being able to detect when it happens, and be able to contend with it if forced into a road racing circumstance.

I would also suggest that there will unequivocally be a need for the AI to have road racing prowess. This will help the AI to deal with day-to-day miniature road racing that happens as part of the mix of human drivers and AI self-driving cars on the roads. It would seem too that humans will relish wanting to see AI self-driving cars that can legally race, doing so in the right situations and with the proper safety precautions undertaken.

Movies like “The Fast and the Furious” have glorified illegal street racing and you can anticipate that our culture will continue to foster that kind of driving approach. If humans are going to be fast and furious, let’s make sure that the AI can be fast when needed, and of course we ought to skip the part whereby the AI becomes furious. Perhaps the movies about AI self-driving cars will be entitled “The Fast and the Faster” and we’ll have eliminated the furious aspect of driving in speed contests.

Copyright 2019 Dr. Lance Eliot

This content is originally posted on AI Trends.