infrastructure · digital · transport · family: the rulebook was written for a world that no longer exists
drone rules that cite standards that don't exist
The FAA built a drone regulation that references standards that don't fully exist yet
Problem statement
The FAA's proposed Part 108 rule (August 2025) establishes a performance-based framework for beyond-visual-line-of-sight (BVLOS) drone operations that replaces traditional airworthiness certification with "airworthiness acceptance" based on industry consensus standards. The problem: those consensus standards don't fully exist. ASTM Committee F38 has published foundational standards (F3478, F3269), but the NPRM itself states that the consensus standards it would rely on "are not yet available." The Department of Transportation's Inspector General found that "FAA's BVLOS operational goals and metrics were difficult for most lead participants to meet" and that "the Agency is not using comprehensive data to inform rulemaking."
Why this matters
Commercial drone delivery, infrastructure inspection, agricultural monitoring, and emergency response have been "just around the corner" for a decade but remain operationally constrained. As of October 2024, the FAA had issued 190 BVLOS waivers under the small-UAS rule (counted from 2020) to 134 distinct operators; the OIG counts BVLOS approvals more broadly (waivers, exemptions and air carrier certificates) and puts them at 26,870 in 2023. The Executive Order's 240-day clock for a final rule (running from June 6, 2025, so roughly February 2026) creates acute urgency — if ASTM cannot finalize the standards the rule will reference, the FAA may default to prescriptive requirements that replicate traditional manned aviation certification overhead, defeating the purpose of the performance-based approach. The EU adopted its U-space regulations in 2021 and has begun implementing them, and the ARC warned that if the FAA, EASA and other authorities "adopt different standards or approaches" for BVLOS, operators and manufacturers "will face barriers to operating abroad and marketing products in foreign markets."
What’s been tried and why it hasn’t worked
The BVLOS Aviation Rulemaking Committee released a 381-page report in March 2022; the DOT OIG counts 140 recommendations in it. ASTM F3478 is a practice for durability-and-reliability flight demonstration programs for low-risk UAS seeking FAA type certification. ASTM F3269 provides a run-time assurance framework for bounding complex function behavior. ASTM published a technical report on autonomy terminology (TR1-EB, 2019), and its AC377 group has drafted a report on regulatory barriers to autonomy. A working-group report inside the ARC report (Phase 1 Working Group 1.3.2, Aircraft and Systems, summarizing a survey of ARC members and experts) proposed Phase 2 frameworks that are "not overly prescriptive," said the level of certification and design assurance used for crewed aviation "is not appropriate for smaller UAS," and named "high-fidelity test requirements that may not be appropriate for smaller UAS" as barriers to entry. Standards such as F3478 were written for the existing type-certification path (14 CFR 21.17(b)), not the new Part 108 framework. The NPRM expects standards bodies to develop the consensus standards after the rule sets the requirements, yet says those standards "are not yet available" — a dependency deadlock. The DOT OIG found the agency "is not using comprehensive data to inform rulemaking," which raises the question of how well the rule is calibrated.
What would unlock progress
Parallel, coordinated development of three interlinked standards: (1) detect-and-avoid system performance requirements validated against representative encounter geometries; (2) command-and-control link reliability requirements based on mission-risk classification; and (3) a modular autonomy assurance framework that extends ASTM F3269 to full mission autonomy. These need to be written to the performance-based philosophy of Part 108, not retrofitted from manned aviation type certification.
Entry points for student teams
A team could develop and test a low-cost detect-and-avoid system for small UAS (using cameras and/or ADS-B receivers) and characterize its performance envelope — detection range, false positive rate, latency — against manned aircraft encounter scenarios. This directly contributes to the performance data ASTM F38 needs. A policy-focused team could analyze the FAA's public list of Part 107 waivers issued (190 BVLOS waivers as of October 2024) to identify which operational conditions and mitigations have been approved, producing an empirical map of regulatory acceptance. Relevant disciplines: aerospace engineering, computer vision, systems engineering, aviation policy.
Genes
Structural cousins: stuck the same way, in other fields
Sources
FAA NPRM, "Normalizing Unmanned Aircraft Systems Beyond Visual Line of Sight Operations," 90 FR 38212, Aug 7, 2025; DOT OIG, "FAA Has Made Progress in Advancing BVLOS Drone Operations but Can Do More To Achieve Program Goals and Improve Data Analysis," Report AV2025034, June 30, 2025; BVLOS ARC Final Report, March 10, 2022; ASTM Committee F38 standards (F3478-20, F3269-21). go to source ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
- The Executive Order's 240-day deadline for a final rule (roughly February 2026) creates one of the tightest standards-development windows in the collection.
- The 2025 Executive Order "Unleashing American Drone Dominance" adds political pressure but does not resolve technical gaps.
- EU/U.S. divergence is a risk the ARC itself flagged ("a lack of uniformity will create uncertainty and impede the development of new and innovative UAS technologies"): the EU's U-space regulations have been in place since 2021 while the FAA's Part 108 is still proposed.
- The dependency deadlock (rule needs standards, standards need rule) is a structural pattern worth tracking across other technology domains.
- The DOT OIG report faults FAA's data practices (not using comprehensive data to inform rulemaking; a validation process that can lead to errors); whether that exposes the final rule to challenge is not something the report says.
Reconciliation 2026-09-30: Whole write-up read against its sources. Four were opened, all on 2026-09-30, all in full: the FAA proposed rule from the Federal Register's plain-text copy (https://www.federalregister.gov/documents/full_text/text/2025/08/07/2025-14992.txt, 180 pages, with the citation record from the Federal Register API); the DOT Inspector General report AV2025034 (https://www.oig.dot.gov/sites/default/files/library-items/FAA%20BVLOS%20Drone%20Operations%20Final%20Report_6.30.2025.pdf, 44 pages); the BVLOS ARC Final Report of March 10, 2022 (https://www.faa.gov/regulations_policies/rulemaking/committees/documents/media/UAS_BVLOS_ARC_FINAL_REPORT_03102022.pdf, 381 pages); and the ASTM store pages for F3478-20 and F3269-21 (https://store.astm.org/f3478-20.html, https://store.astm.org/f3269-21.html, scope and significance sections only, the standards themselves are paid). (1) The Source line cited the proposed rule as "90 FR 64766"; the Federal Register record gives 90 FR 38212, and the OIG report's title was paraphrased; both now carry the strings from the records, and the ARC report date is given as March 10, 2022. (2) The text said "critical gaps remain in detect-and-avoid performance standards, command-and-control link reliability standards, and comprehensive autonomous flight standards." None of the sources lists those three gaps; the ARC report and the proposed rule both cite an existing ASTM detect-and-avoid performance specification (F3442). What the proposed rule does say is that "the consensus standards are not yet available"; the sentence now says that. (3) "The FAA's own Inspector General" is the Department of Transportation's Inspector General; corrected in two places. (4) "Only 190 BVLOS waivers total": the OIG report gives 190 BVLOS waivers issued from 2020 through October 2024 under the small-UAS rule, to 134 distinct operators, and separately counts BVLOS approvals through waivers, exemptions and air carrier certificates at 26,870 in 2023; the sentence now gives both. (5) The "February 2026 deadline": no source names that month; the proposed rule quotes Executive Order 14307 (June 6, 2025) directing a final rule "within 240 days," which the text now says, with February 2026 as the arithmetic. (6) The claim that the EU's U-space regulation "is proceeding with its own standards framework" and "would fragment a global drone industry" is not in the sources in those words; the proposed rule says EASA adopted three regulations in 2021 and has begun implementing U-space, and the ARC report warns that if the FAA, EASA and other authorities "adopt different standards or approaches" for BVLOS, operators and manufacturers "will face barriers to operating abroad and marketing products in foreign markets" and "a lack of uniformity will create uncertainty and impede the development of new and innovative UAS technologies." The sentence, and the matching Source Notes bullet, now say that. (7) "70 recommendations": the ARC report does not state a count, and the OIG report says the ARC report "contained 140 recommendations"; the text now gives the OIG's count. (8) "ASTM F3478 was published as a 'pillar' standard for UAS certification and recognized by EASA": neither the word "pillar" nor any EASA recognition appears in the sources. The ASTM page describes F3478 as a practice for durability-and-reliability flight demonstration programs for low-risk UAS seeking FAA type certification under 14 CFR 21.17(b); the text now says that, and the EASA-recognition bullet in Source Notes was reworded. (9) "The FAA's Strategic Advisory Committee (AC377)": AC377 is an ASTM committee, not an FAA one; the ARC report lists ASTM's technical report on autonomy terminology (TR1-EB, 2019) and a draft AC377 report on regulatory barriers to autonomy. The sentence now says that. (10) The quoted sentence "many airworthiness design and test requirements from the TC process are overly prescriptive and do not adhere to the performance-based approach" is in none of the sources. It was replaced with what the ARC report does contain, in the Phase 1 report of Working Group 1.3.2 (Aircraft and Systems), which summarizes a survey of ARC members and experts and proposes Phase 2 tasking: frameworks should be "not overly prescriptive," the level of certification and design assurance used for crewed aviation "is not appropriate for smaller UAS," and "high-fidelity test requirements that may not be appropriate for smaller UAS" are barriers to entry. (11) "ASTM standards were designed for the existing regulatory paradigm" now names F3478 and its type-certification scope, which is what the ASTM page supports. (12) "Meaning the rule itself may be poorly calibrated" and the Source Notes speculation that the rule "may face legal challenge on evidentiary grounds" are not in the OIG report; both were softened. (13) The entry point's "BVLOS waiver database (190 waivers, publicly available)" now points to what exists: the FAA's public list of Part 107 waivers issued, which the proposed rule cites and which was opened on 2026-09-30 (https://www.faa.gov/uas/commercial_operators/part_107_waivers/waivers_issued). Confirmed and left as written: Part 108 and "airworthiness acceptance" based on industry consensus standards (proposed rule); the two OIG quotations; the 381-page length of the ARC report (page count of the PDF) and its March 2022 date; F3269 as a run-time assurance practice for bounding complex functions; Executive Order 14307 "Unleashing American Drone Dominance" (2025); the FAA's request for input on whether it should write its own means of compliance in lieu of consensus standards. Not checked: the three-standard proposal in "What Would Unlock Progress" and the "just around the corner for a decade" framing are the write-up's own analysis, not sourced claims; the earlier Retagged bullet's phrase "standards Part 108 references but that do not exist" is left as it stands (it is a tagging note, not a sourced claim). No title changed. Genome tags untouched. A second reader went over these repairs on the same day, opening the same sources independently (all four, plus the FAA Part 107 waivers page); the ARC report's passage on FAA/EASA divergence, which the first pass had not found, was added under (6), and the attribution of the "not overly prescriptive" passage under (10) was narrowed to the working-group report it comes from; the other eleven repairs were confirmed in the sources and stand as written.