Beyond Line of Sight: How Part 108 Could Transform Commercial Drones

In September of 2024, Hurricane Helene tore through western North Carolina, washing away roads, severing communications, and isolating entire communities for days. Emergency responders increasingly turned to drones to locate survivors, assess damaged infrastructure, and identify passable routes without putting rescue teams in additional danger. The use of such drones has radically changed the ability of first responders to operate in disasters. Previously, such work required ground crews or costly crewed aircraft, meaning the ability to survey such areas was slower and more limited. Now, first responders can perform such tasks rapidly and repeatedly from the air, providing actionable information to guide when and how they deploy in challenging environments. In the case of the response to Hurricane Helene, drones quickly assessed which areas were inaccessible due to road damage and identified dangerous conditions such as landslides.

The deployment of such drones transformed responders’ ability to quickly survey challenging areas. Yet federal regulations handicapped many of the flights that could have provided the most value—by covering long stretches of flooded highways, delivering supplies to isolated communities, or surveying multiple disaster sites from a single launch point—with requirements that made operators keep drones within their direct visual line of sight.

Under the Federal Aviation Administration’s (FAA) Part 107 rules, most commercial drones cannot fly “beyond visual line of sight” (BVLOS) without obtaining an individual waiver. In practice, that means a drone pilot, or a network of visual observers, must maintain uninterrupted visual contact with the aircraft throughout its flight. While appropriate for low-risk operations envisioned when Part 107 was introduced in 2016, these restrictions become a significant limitation during disasters, when the areas requiring assistance and inspection may stretch for dozens of miles.

Part 108 is the FAA’s proposed framework for routine BVLOS operations, establishing a separate regulatory pathway that complements Part 107 for more complex missions. By allowing qualified operators to conduct safe, scalable BVLOS missions without relying on case-specific waivers, the regulation has the potential to transform emergency response, infrastructure inspection, agriculture, logistics, and other critical industries. The FAA reopened a targeted comment window in January 2026 on right-of-way, ADS-B, and detect-and-avoid provisions, and a draft rule was submitted to the Office of Information and Regulatory Affairs in July 2026, where it remains under review.

The stakes of the FAA’s decisions go far beyond drones. This is an early test of the ability of the United States to develop a regulatory framework that enables autonomous systems to operate in shared infrastructure at scale while preserving a high level of safety.

Benefits of Part 108

Recognizing that Part 107 could no longer support the rapidly expanding commercial drone industry, the FAA convened the BVLOS Aviation Rulemaking Committee in June 2021. Comprising representatives from Amazon Prime Air, Wing, and other aviation organizations, the committee produced 70 recommendations that ultimately formed the basis for the proposed Part 108. Congress reinforced the urgency of this effort through the FAA Reauthorization Act of 2024, directing the agency to develop a permanent regulatory framework for routine BVLOS operations.

The philosophy underpinning Part 108 marks a meaningful departure from the prescriptive approach of Part 107. Rather than tying permissions to specific locations or requiring one-off waivers, it is designed to establish a scalable certification framework in which operators demonstrate their systems meet defined safety standards. Once approved, operators would be permitted to conduct BVLOS flights across a broad range of environments without repeatedly seeking FAA authorization. The result is a regulatory system designed to accommodate commercial operations at scale rather than isolated demonstrations.

Critically, the rule also recalibrates how safety responsibility is assigned, shifting the compliance burden from individual pilots to the operator organization and better reflecting how modern, corporate drone programs are actually managed. Under Part 108 operations, pilots would no longer need to obtain a traditional airman certificate; instead, training responsibility would move to the operator. Operators would be required to employ operations supervisors to oversee overall safety and flight coordinators to monitor individual flights. The strictness of safety requirements would vary based on risk level: A drone flying over a large crowd would face more stringent rules than one operating over an open terrain, and in the highest-risk scenarios, drones would be mandated to detect all other aircraft flying within the vicinity. The rule would significantly broaden where drones could legally fly, allowing operations across Class B, C, D, E, and G airspace, including in the vicinity of airports. It would also permit drones to fly within 50 feet of a structure as well as flights over people, opening the door to urban drone services.

Part 108 should not be viewed in isolation, but rather as part of a broader effort to develop the digital infrastructure needed for autonomous aviation at scale. The proposed Part 146 would create a system for FAA oversight of automated data service providers and other entities that support aircraft operations; as the FAA explains, “the ability to operate BVLOS is predicated on the fidelity and assurance of the data.” This could operate as a new digital infrastructure layer that transforms the architecture of safety for low-altitude operations: Rather than focusing solely on the reliability of individual aircraft, aviation safety would depend on a distributed system of software, networks, humans, and automated services that connect, coordinate, and support them. Developing such a system will require infrastructure that is secure, interoperable, and resilient. Part 108 therefore has the potential to serve as a model for other autonomous systems, which will equally depend on having such distributed systems to achieve safety and scale.

Current Debates and Underlying Issues

While Part 107 remains appropriate for less complex operations, there is broad agreement that advanced operations require a more scalable regulatory framework. At the same time, stakeholders remain divided over how much risk the FAA should tolerate as autonomous aviation expands.

The most significant debate concerns detect-and-avoid (DAA) requirements. The FAA proposes requiring drones to detect both cooperative aircraft, which broadcast their positions, and noncooperative aircraft that do not—but only for operations conducted in Class B or Class C airspace, or over Category 5 operations, the most densely populated areas. For Categories 1 through 4 and operations outside those airspace classes, no noncooperative DAA requirement is proposed. Some drone manufacturers argue that even this limited requirement imposes an unrealistic technical burden and instead advocate requiring more crewed aircraft to carry ADS-B or other electronic conspicuity systems. General aviation organizations, however, contend that many legacy aircraft cannot reasonably be expected to install equipment and argue that drone operators entering the national airspace system should bear the responsibility for avoiding collisions.

This reveals a fundamental problem with the proposed framework: Ground risk and air-collision risk are distinct safety problems. Population density is useful for assessing the potential consequences if a drone crashes into the ground, but it is less relevant to determining whether a drone is likely to collide with another aircraft. Sparsely populated areas may see low-altitude aircraft operate frequently for certain use cases, while some densely populated areas may not have a high concentration of such operations. Thus, tightly pairing DAA requirements with population density risks both overregulation—imposing more stringent requirements than are justified in some areas—and underregulation—imposing less rigorous protections in other areas where there is significant risk.

Ground risk should affect how to protect people located underneath a drone, but population density should not be used as a proxy for air-collision risk. Electronic conspicuity is an important element in the safety architecture, with broad adoption helping to reduce the risk of airborne collisions. At the same time, it should constitute one layer of protection in an overall, multilayered safety system, rather than becoming the sole factor in evaluating BVLOS operations.

Achieving a safe and resilient safety architecture requires a multilayered approach to aviation safety, rather than relying on a single technology or proxy for how airspace users will behave. Key layers in such an architecture include strategic deconfliction to reduce the risk that aircraft are in paths that may collide, electronic conspicuity to make it easier to detect and avoid other aircraft, and independent DAA capabilities to provide additional safety assurance when other layers in the system fail. Requirements for each layer should be performance based and technology neutral, enabling operators to quickly and seamlessly deploy new, innovative solutions.

Oversight presents an equally important challenge. The Part 108 proposal offers limited guidance on maintenance standards comparable to Part 145 repair station certification, which is a well-established requirement in crewed aviation. More broadly, the question of how the FAA will police the new entrants looms large. As drone registrations grow, so too do disruptions at outdoor public gatherings. During its 2023 season, the National Football League recorded over 2,800 drones breaching restricted airspace near its stadiums, up from just a dozen logged in 2017, and drone sightings have even been reported by the U.S. military over its facilities. Until the Safer Skies Act was enacted as part of the 2026 National Defense Authorization Act, only a few federal agencies had the legal authority to jam, intercept, or disable rogue drones. Whether state and local authorities will be able to effectively exercise their newly granted permissions remains an open question.

Finally, the regulation has generated debate over access to hardware. The proposal would create what critics describe as a backdoor ban on widely used drones, including those manufactured by Chinese companies such as DJI, by requiring bilateral agreements between the United States and the drone’s country of origin. This raises difficult trade-offs between advancing security and supply chain integrity and increasing costs or limiting U.S. businesses’ access to highly capable hardware. The broader goal should be to leverage Part 108 to foster a trusted drone market in which U.S. and allied suppliers can compete, innovate, and support deployment at scale.

Recommendations

Part 108 is a necessary expansion of the FAA’s commercial drone framework. There is tremendous potential in moving commercial drone operations from one-off waivers to a scalable BVLOS framework, unlocking major benefits across logistics, infrastructure inspection, and emergency response. The final rule, however, should avoid creating a safety architecture that depends too heavily on assumptions about legacy aircraft or technologies that have yet to mature.

To secure U.S. leadership in autonomous aviation, the FAA should adopt a regulatory framework based on performance rather than prescription. Safety standards should reflect real-world operating environments, incorporate multiple layers of redundancy, and remain technology agnostic so that operators can satisfy regulatory requirements through onboard sensors, ground-based infrastructure, or future innovations rather than a single mandated solution. The choice is not between safety and scale; on the contrary, providing confidence in safety is a prerequisite for achieving scale.

The FAA should mandate independent DAA capabilities wherever BVLOS operations present a meaningful risk of encountering noncooperative crewed aircraft, rather than determining those requirements primarily through population density. It should pair those requirements with a long-term strategy to improve electronic conspicuity across the broader aviation ecosystem. Safety should not rest entirely on either drone operators or crewed aircraft, but instead on complementary protections that reduce the potential for single points of failure.

Finally, regulatory expansion must be accompanied by modernization of oversight. Allowing routine BVLOS operations without sufficient inspection capacity, maintenance standards, or enforcement authority risks undermining public confidence in the technology.

As the most consequential overhaul of commercial drone regulation in nearly a decade, Part 108 will shape not only how drones operate today, but whether the United States establishes the regulatory foundation necessary to lead the next generation of autonomous aviation. If the FAA gets that foundation right, it can create a model that extends well beyond aviation.

Matt Pearl is director of the Strategic Technologies Program at the Center for Strategic and International Studies (CSIS) in Washington, D.C. Kuhu Badgi is program coordinator and research assistant in the Strategic Technologies Program at CSIS.