Nine months ago, the Federal Aviation Administration (FAA) published the rule that was supposed to make routine beyond-visual-line-of-sight (BVLOS) drone operations a reality. The final version was due in March 2026. It is now September, and Part 108 still sits with the Office of Information and Regulatory Affairs (OIRA), where it has been under review since July 10 — while the industry has crossed 65 million flights without it.
The industry is scaling anyway. AirData passed 65 million drone flights in August 2026, covering 850,000 aircraft and 470,000 pilots, at a pace of roughly a million flights a month.
The question is no longer whether BVLOS gets certified. It is who gets certified first — and whether the US rule's complexity pushes early scale-ups toward India's end-2026 BVLOS path or the EU's U-space instead.
Part 108 matters because it would replace today's site-specific waiver system with performance-based certification. Operators get approved once, then run routine flights across a defined area. The drone economy's growth assumptions are built on that shift.
How BVLOS Became the Industry's Longest Wait
Under Part 107, every beyond-visual-line-of-sight operation still needs its own waiver. A national pipeline or powerline inspection program can require twenty or more separate approvals, each with its own safety case, geography, and operating conditions. Nothing about the system scales.
The reform path has been running for years. The 2024 FAA Reauthorization Act made BVLOS rulemaking a statutory mandate. In July 2024 the agency issued its first automated data service provider (ADSP) letters of acceptance to Zipline and Wing. In June 2025 an executive order set a 240-day clock for the final rule. The Notice of Proposed Rulemaking (NPRM), covering Part 108 for operators and Part 146 for traffic-management services, was published on August 7, 2025.
The comment period closed October 6. A narrow reopening on electronic conspicuity ran through February 11, 2026. The March target passed without publication. As of late August 2026, the final rule is still with the office — the last step before it appears in the Federal Register.
The gap between the industry's readiness and the regulator's pace is the whole story of the sector in 2026.
The Case for Routine BVLOS
The optimists' first point: when the rule lands, it is designed for autonomous systems, not adapted from rules written for human pilots. Part 108 is performance-based: detect-and-avoid requirements keyed to what the aircraft can actually do, an organizational operator model, and a shift of responsibility from an individual remote pilot to an operations supervisor inside the company.
The scale evidence is already on the books. AirData crossed 50 million tracked flights in May 2025 and 65 million by August 2026 — 15 million flights in fifteen months, roughly a million a month. The fleet it tracks has grown to 850,000 drones and 470,000 pilots. Much of that growth is organizations running drones as routine infrastructure, not as pilot programs.
The competitive argument is sharper than the technical one. India's Civil Aviation Ministry said in August 2026 that BVLOS certification would land by the end of the year. Canada standardized BVLOS rules in 2023. The EU has been implementing U-space since 2021. US operators are already flying BVLOS abroad — in Italy, the UK, Rwanda, and Japan — because the domestic waiver path is the slowest part of their business.
Companies that align their operations with Part 108's requirements now capture a market that has been artificially constrained for nearly a decade. That is the bull case in one line.
The Case That the Wait Isn't Over
The skeptic's answer is that the wait is the point. The final rule is not published. Practitioner expectations, tracked through July 2026, range from imminent to 2027–2028 planning assumptions. The industry has heard "soon" from regulators for the better part of a decade, and each time the deadline slipped for structural reasons.
Implementation will not be a switch-flip either. The proposed framework leans on ADSPs — third-party traffic-management services with uncertain pricing — plus operational-area approvals and detect-and-avoid certification. Each of those steps can recreate the very bottleneck the rule was meant to remove, this time as a certification queue instead of a waiver queue.
Opposition is organized. Pilot associations and agricultural aviation groups object to the right-of-way provisions that would make unequipped manned aircraft yield to drones. News organizations have raised First Amendment concerns about urban density limits and advance flight-planning requirements. Even if the final rule publishes this fall, contested compliance detail moves into advisory circulars with their own comment cycles.
A concentration risk sits underneath all of it. Large operators with money to build ADSP relationships and safety cases will clear the certification queue first. Smaller operators, already squeezed by the waiver system, may find the new framework widens the gap rather than closes it.
Three Clocks: FAA, India, and the EU
Watch it as a race between three regulatory clocks, not one.
| Jurisdiction | Status | Target | What it unlocks |
|---|---|---|---|
| United States | ◐ Final rule pending at OIRA since July 10, 2026 | Unpublished; expectations from imminent to 2027–28 | ✔ National certification instead of per-site waivers |
| India | ◐ Certification framework in development | BVLOS certification by end-2026 | ✔ Scaled commercial BVLOS; manufacturing hub push |
| European Union | ◐ U-space implementation since 2021 | Gradual integration | ◐ Cooperative airspace for manned and unmanned |
The delay is not neutral.
Every month the US rule stays at OIRA is a month in which certification-driven scale-up happens elsewhere. The framing that survives both positions is simple: the market does not wait for the certificate.
The Stakes: a Market Waiting on a Certificate
What Part 108 actually gates is not technology. It is cost. BVLOS cuts the single biggest operating expense in drone services: the person on the ground watching the aircraft. Remove that observer and a drone team can run multiple aircraft from one desk, cover a pipeline across three states without a convoy of vans, and scale inspection and delivery capacity without hiring pilots for every site.
That is why the certification race matters to investors. The companies that clear the queue first get the margin advantage, because their operating cost per flight drops the moment the observer requirement disappears. AirData's numbers suggest the demand is already there: a million flights a month, mostly under the waiver system that everyone agrees is too slow.
The US has a structural advantage it is spending. It wrote the waiver playbook, hosts the largest commercial drone fleet, and its aerospace regulator has more BVLOS operational data than any other agency on earth. Every month of delay converts some of that advantage into a footrace with India, which has fewer installed drones but a faster regulatory clock and a government that treats drone manufacturing as an export priority.
Neither position in this debate predicts the other away. As we wrote in September, Scout AI is betting $100 million on an autonomy brain that commands no hardware of its own — the same software-first logic the organizational-operator model assumes. The certification debate sits on top of that bet. The bull case rests on the rule being a genuine unlock. The bear case rests on implementation turning the unlock into a new bottleneck. Both agree on the operating truth: the market is scaling now, and certification determines who captures that scale at the best margin.