Delivery drones are coming.
Drone naysayers, I've got bad news. You've lost the war. Drones have already become indispensable in a handful of sectors (infrastructure and agricultural inspection, and search and rescue, to name a few). When home drone deliveries arrive, which they will in the next few years, those buzzy bastards will be everywhere.
Welcome to the age of the commercial drone.
So let's talk inevitabilities. What happens when a guy out for a jog gets brained by a delivery drone? Or when a McDonald's flyer douses a woman with scalding coffee? Even as regulatory hurdles fall, liability will be the major limiting factor to drone ubiquity.
There are already a handful of companies that offer commercial drone insurance. I got a chance to talk with Mike Kelly, the cautiously-titled Media Risk Control Manager at
Prosight Specialty Insurance, which insures businesses that use commercial drones. It was an illuminating conversation. Turns out even small companies may offer drone services. What are the current rules when it comes to flying commercial drones over people?
The FAA has approved UAV flight over people, but only in closely regulated circumstances, where everyone over which the drone is flying is working on the project the drone is covering. An example of this is closed-set filming for motion pictures. Drones are used frequently in these projects and a drone operator with the proper 333 exemption may fly the drone over cast and crew - but not the general public.
The Federal Aviation Administration (FAA) is
establishing an aviation rulemaking committee with industry stakeholders to develop recommendations for a regulatory framework that would allow certain UAS to be operated over people who are not directly involved in the operation of the aircraft. The committee will begin its work in March and issue its final report to the FAA on April 1.
UAVs are already delivering packages in other countries. Switzerland's postal service has begun testing parcel deliveries by unmanned drones, although widespread use of the UAVs is not likely to be implemented for another five years. Medical supplies have been delivered to remote areas in Africa for several years. In July 2015, the
FAA approved the first such use of a drone within the United States, to deliver medicine to a rural Virginia medical clinic. Another creative delivery niche is also being fine-tuned in the United States, Ireland, Britain, Australia and Canada, where weapons and drugs have been dropped in to prison yards by drones. What are some of the biggest issues faced by companies looking to implement drones? There are already hundreds of companies using drones every day. These industries include aerial photography, real estate, agriculture, search and rescue, mining, and closed-set filming, industrial refinery and manufacturer plant inspection, utility system inspections, private security, pipeline inspection, wildlife and forestry monitoring, construction site inspection and cell tower inspection. The biggest challenge is to ensure they are working with a drone operator who not only has his 333 exemption, but that the exemption specifically approves the specific intended operation. You can be approved by the FAA for aerial photography but not for closed set filming or inspection of roof mounted ventilation equipment. And if your drone operator has an accident while conducting a non-approved operation, it's a significant liability exposure where it would be easy to allege negligence. So it's very important that a company looking to use a drone do their own investigation on the FAA site to ensure the operator is approved for the specific intended task. Anyone looking to deploy drones in to the public sector must be aware of the importance of a fail-safe system to minimize the potential for damage or injury in event of a UAV failure. Technology presently exists that will allow a parachute to deploy in event of a failure, but if the disabled drone parachutes on to a busy freeway, there could still be a problem. And when the UAV is on the ground, there is potential for theft or vandalism. Technology also exits that will remotely disable the UAV in event of theft. But the loss of a drone, even if it's disabled could be very costly. Drones are not the silver bullet for product distribution. In their current form, drones must frequently recharge their batteries, and this, together with the mandatory and practical weight restrictions on the weight of a package to be delivered will likely require a drone to return to base after each delivery, meaning that a drone would not be able to match the 200 - 300 deliveries per day accomplished by FedEx or UPS truck in urban and suburban areas. One of the most important concepts in the world of liability and litigation is the concept of foreseeability. That is, was the incident (personal injury or property damage) foreseeable? It doesn't matter whether the act that resulted in the incident was against recognized common sense or formal safety standards. If an action that could result in personal injury or property damage is foreseeable, then the company using the drone may be held accountable for preventing it. So the liability related to flying a drone over people, where it could possibly fail and crash, injure people with spinning blades while delivering a package or drop a package on to a pet in the back yard doesn't appear to ever be 100% manageable. For businesses looking to invest in drones, what are some of the most prominent insurance concerns to be wary of? It's all about understanding the regulatory environment - you've got to have a clear understanding of what's legal and what's not. The first step has got to be the FAA's 333 exemption. And it's critical to ensure that you request the exemption for the exact activity you plan. Many drone operators have gone through the process of obtaining a 333 exemption only to find that they did not request to be approved for every specific foreseeable use, and if the 333 does not specifically permit an activity, the operator is prohibited from conducting that activity. I think it's an accurate statement to say that the majority of commercial drone operators are operating without a 333 exemption. As of today there are only 3927 drone operators with a 333 exemption, and that's across all applicable industries. But there are actually a few insurance companies that say they will issue a policy to those companies, but the reality is that it's against the law to insure an illegal activity, so in the event of a big loss, the insurance company can easily deny coverage. So - to properly control the liability exposure, you've got to ensure you're operating legally - again - conforming to not only the FAA requirements, but also to any regional requirements as well. And it's critical to be aware of all applicable requirements. Many states, counties, municipalities, federal lands and individual cities have laws and ordinances directed specifically at the use of UAVs, and these rules may be far more stringent than the requirements one had to implement to obtain a 333 exemption. How do pilot education programs and credentials come into play? Does the experience/knowledge of the pilot play a part in liability? Anybody who has every applied for a job has had to submit a resume, and the more comprehensive the resume, the more confident we are that the applicant has the capabilities we seek. It's the same here. When a drone operator come s to me seeking insurance coverage, the first thing I look for is their 333 exemption, but just like people with a college degree but no actual experience, I look beyond the exemption to see if they have any additional education or experience. In the drone industry, however, most of the education and experience is informal, just going out and flying the drone and trying not to crash it in to anything. There are hundreds of very competent film makers who are expert drone operators that have yet to obtain their 333 exemption, so I not only look for formal education, but also evidence of substantial practical experience. One the front end, the amount of education and experience makes me more comfortable with a particular operator, and in the event of an incident, the operator has a more substantial defense if he can show that he has a formal education, an excellent safety record and was operating within the applicable regulations. There are presently hundreds of institutions that offer courses and certificates (not FAA 333 exemptions) for drone operations. These all support the credibility and competence of the operator should there be a claim to the contrary. What are some things you should keep top of mind when it comes to drone maintenance? Does that play a part in liability? While the best way to avoid liability is to never have a claim or an incident, in a practical sense, avoiding liability is all about being able to prove you're not negligent. And one certain way to appear to be negligent is to fail to maintain, or fail to be able to prove you've maintained your airframe. It's not enough to do it, you have to be able to prove it. And when a potential client comes to me, it's one of the first things we evaluate - prove to me how you maintain your airframes. Specific maintenance requirement s are conspicuously absent from the current FAA regulations. 333 exemptions do not propose any requirements for airworthiness or for inspections by a certificated repairman, but the fact that they don't tell you exactly what you have to do doesn't mean you don't have to do it. In the world of liability, the mandatory standards should always be considered the minimum standard. Doing more than you have to is a better defense than doing only the minimum. So in this case, it's absolutely imperative that the operator follow the manufacturer's care and maintenance instructions. Manufacturers are so acutely aware of their product liability exposures that it's a reasonable assumption that each commercially produced airframe will have comprehensive care and maintenance instructions. Just as is the case with manned aircraft, the insurance world will likely have a significant influence when maintenance related requirements are developed. If the FAA allows drone delivery services, what will it mean from a risk and liability perspective? What we're trying to avoid is personal injury and property damage, and whenever you've got stuff in the air, there a potential for it to fall. And if it falls, there's potential for personal injury or property damage. But manned aircraft don't regularly fall out of the sky so there's no reason to believe that unmanned aircraft would be more prone to crashes, assuming there are regulations, as there are with manned aircraft, to ensure the competence of the operators and the dependably of the airframe. The FAA is responsible for the safety of U.S. airspace from the ground up, so if there comes a time or a place where drones can deliver stuff, it means that the process has been vetted and determined to safe. Having said that, just because it's legal doesn't necessarily mean there is no risk, and that's where my job comes in. I identify the exposures and evaluate the controls and make the determination as to whether a loss is likely. I can't see a time when dropping packages on to someone's front porch, or landing a drone in someone's back yard will ever be free from risk. There are too many variables. But medical supplies and specimens have been safely delivered by drones in Switzerland, Haiti and the Dominican Republic, since 2011. And since nobody can conduct this type of operation without insurance, it will be up to the drone operators to develop and implement FAA approved protocols to minimize the potential for accidents, and it will be up to the guys that do what I do to determine whether these protocols are sufficient to minimize the potential for a loss. We insure many drone operators now working in a wide variety of industries, but we look at each one on a case by case basis to make sure we understand the nature of the operation, and that will likely be the way it will go for the foreseeable future. There isn't yet enough continuity and loss history within any particular niche for us to approve a particular type of use without looking at every element of every operation. It all has to start with the FAA approving deliveries, and the FAA is likely to initially only approve very specific types of deliveries to ensure the safety of the airspace. So I think deliveries from the roof of a hospital to the roof of a lab are more likely to be approved before Dominoes gets to deliver pizzas to your house while you and your friends and your dog stand out on the driveway to watch it land. But companies like ProSight will be at the forefront of that industry because we have the experience and the expertise to understand the technology. We already insure many drone operators so once the FAA approves deliveries, I'm sure we'll be able to provide converage for the operators who conform to the applicable standards. But having said that, the laws governing the use of drones varies from state to state, and sometimes from county to county, so just because a drone operator conforms to FAA requirements, doesn't mean they will be able to operate everywhere, and that's where there's significant liability exposure, to make sure the operation complies with all mandatory standards, and from a liability standpoint, it's also critical to understand the applicable voluntary standards that may apply, because if there's an accident, the plaintiff's attorney can allege negligence if the operator didn't comply with the applicable best practices, whether they're mandatory of not. How will drones flying in the air delivering packages increase the risk of accidents and collisions? That's really the trick, isn't it, and that's why current drone regulations require drones only to be operated during daylight in line of sight. Once an unmanned aircraft goes beyond line of sight, it's far more difficult to guarantee it won't run in to anything. The technology for recognizing adjacent air traffic and taking appropriate action to avoid a collision exists, but it's not yet been integrated in to commercially available drones. And that concern that must be solved before we agree to insure beyond line-of-sight UAV flights through potentially crowded airspace. One of the companies leading the industry in this area has developed a system whereby the drones will only fly to a proprietary, sensor-equipped landing pad, thereby eliminating the potential for landing in an unapproved spot. While this would preclude delivery of retail products to consumers, it may likely be an important consideration in the approval process for the use of drones to deliver objects from one established base to another. Accidents can also occur while the drone is on the ground, since there can be as many as 16 spinning blades that have proven able to cause serious injury if touched, so while this is an exposure the manufacturers will have to address, it's also an important consideration for anyone landing a drone where unauthorized individuals may be able to touch the machine. The technology exists for the drone to automatically stop its propellers so the contents can be retrieved -- but in a residential delivery scenario, this would not only require the drone, with multiple spinning blades to come to a landing in the close proximity of anyone nearby, but it would also necessitate some interaction by the consumer to give the "all clear" to the UAV before takeoff - a very unpredictable scenario. The other delivery option is to allow packages to drop so the UAV does not have to land. The potential for a negative incident here is obvious.