In the video above, one of Kodiak's trucks can be seen traveling down the highway at 65 miles per hour in the right-most lane. An occupant inside the truck (but not at the wheel) cuts a cable to simulate a sudden loss of communication between the truck and its main computer. That's when Kodiak's Actuation Control Engine kicks in. ACE is Kodiak's fallback system which enables the truck to safely pull over to the side of the road when it notices a crucial system failure. In this circumstance, ACE recognizes the interruption from the cut cable and plots a lane path ahead of the truck to move onto the shoulder and begins to brake.

When the number of users for a predictive model grows, it is expected (albeit often wrongly) that the machine learning powered systems will automatically scale to keep up with this growth. If the system fails to scale, processing requirements may outpace performance. Using an example from a LinkedIn article, a sample recommender system fails to recommend the desired list of products or services in a timely manner, which means the customer does not receive the product or service recommendations at the time of purchase. Though developing a scalable system can pose a serious challenge, shying away from building a scalable system can become a bigger problem and can result in lost customers or unrealized revenue. During scaling, many technical problems like workload issues, memory representation, framework restrictions, resource use vs. performance, and others can surface and stall the production.

Though nearly a third of American adults still report experiencing fear of flying, the majority of us implicitly trust that airlines will get us from point A to point B and back again in one piece. Indeed, few factors are as crucial in the transportation industry as trust. If we didn't believe in the safety of a train, plane or automobile, no one would ever travel that way. Planes, however, still have pilots, and trains usually have conductors, and most likely will for the near future. Interestingly, the introduction of autopilot in commercial aviation – effectively rendering pilot control for large parts of the flight unnecessary – did not seem to bother most passengers, (similar to modern conductor-less locomotive systems).

Kaushik Raghu, Senior Staff Engineer at Audi, is reflected in the passenger side visor mirror while demonstrating an Audi self driving vehicle on I-395 expressway in Arlington, Va., Friday, July 15, 2016. Experts say the development of self-driving cars over the coming decade depends on an unreliable assumption by most automakers: that the humans in them will be ready to step in and take control if the car's systems fail. Experience with automation in other modes of transportation suggests that strategy will lead to more deaths like that of a Florida Tesla driver in May. Kaushik Raghu, Senior Staff Engineer at Audi, is reflected in the passenger side visor mirror while demonstrating an Audi self driving vehicle on I-395 expressway in Arlington, Va., Friday, July 15, 2016. Experts say the development of self-driving cars over the coming decade depends on an unreliable assumption by most automakers: that the humans in them will be ready to step in and take control if the car's systems fail. Experience with automation in other modes of transportation suggests that strategy will lead to more deaths like that of a Florida Tesla driver in May.

Ford recently drove an autonomous Fusion through the Arizona night, relying entirely on the car's laser radar system to plot its course. Ford engineers working on the company's autonomous car technology recently succeeded in making a self-driving Ford Fusion lap its Arizona Proving Grounds in complete darkness, using laser radar, or Lidar, as its guide. While Lidar is a standard feature of most self-driving car gadgetry - along with traditional radar and cameras - the test was meant to specifically highlight how an autonomous vehicle might operate in a situation that many humans find nerve-wracking. Most self-driving cars are tested during daylight hours so all of their electronic systems can contribution to the process. As autonomous vehicles creep into our automotive future, the biggest hurdle for automakers won't be scientific but rather psychological.