L-Re · Incidental Research

The Clearance, Not the Missile

# The Clearance, Not the Missile

*First Combat Hellfire and Split CIA/USAF/CENTCOM Control*

**Prepared as a capability and political-economy analysis**  
Current as of 19 August 2026

---

> **Research premise.** On 7 October 2001 the host nation approved weapons-carrying flights, Operation Enduring Freedom opened, and a Predator fired the first combat Hellfire — at a security vehicle near a compound associated with Mullah Omar, after Central Command delayed a shot on the building itself. The trust failure is remote-kill command and control, not the missile.

## Abstract

**Research premise.** A precision missile that hits its aimpoint on the first night of a war is easy to misread as a weapons story. The public record of October 2001 does not support that reading. The same day the host nation approved weapons-carrying flights and Operation Enduring Freedom opened, air vehicle 3034 held a firing position on a convoy associated with Taliban leader Mullah Omar near Kandahar and requested clearance. Authorization from United States Central Command was delayed; Omar entered a structure; concerns about incidental casualties produced a decision to strike a nearby security vehicle rather than the building. Mid-month Pentagon press then confirmed what the shot had already exposed: a joint Central Intelligence Agency and Air Force operation with split control — Agency day-to-day ownership of the hunt, Air Force flying from the United States after takeoff, Agency backup if the satellite link dropped, and a combatant-command release that sat in Florida while the sensor-shooter loitered over Afghanistan. This paper treats that month as a capability and political-economy problem. The Hellfire had already been test-fired from the same airframe in the Nevada desert. The wartime constraint was the Control Fabric: who may Sense, who may Move, and who may Act, across a satellite path, an agency seam, and a combatant-command lawyered release. Exact casualty counts on the vehicle and the unpublished rules of engagement remain unknown. They are not safe defaults. The honest unit of analysis is clearance latency against a fleeting high-value individual, not the missile that finally left the rail.

_Keywords: armed unmanned aircraft, Predator, Hellfire, command and control, Central Command, Central Intelligence Agency, Control Fabric, Sense-Move-Act, product assurance, political economy of remote kill_

## 1. Introduction: The Apparent Puzzle

A combat-aviation headline reduces 7 October 2001 to a first. An unmanned aircraft fired a Hellfire in war. A reader who treats the weapon as the event expects a story about accuracy, fusing, and the conversion of a reconnaissance airframe into a shooter. The contemporaneous and museum records of that night do not make that calculation. They describe a crew that already held a firing position, a request for clearance, a delay at Central Command, a high-value individual who walked into a building, and a substitute aimpoint chosen to flush him rather than to hit the structure he had entered.1

The disparity is easiest to misunderstand when the month is read as a munitions debut. Under that reading, February’s Nevada tests are the technical climax and October is the operational proof. The National Air and Space Museum’s object record for MQ-1L Predator A, inventory A20040180000, does identify air vehicle 3034 as the first unmanned aircraft to fire Hellfire in combat. It also records that the first combat shot was not the shot the crew had asked to take.2 That is not a footnote. It is the analytic object.

This distinction matters. Capability analysis can target what a system can Sense, how it Moves, and whether it can Act. Product assurance adds a fourth question: whether the command path that joins those three functions is owned, timed, and logged tightly enough that a fleeting target remains a target. October 2001 answers the first three more cleanly than the fourth. The sensor worked. The satellite path worked well enough to put a crew on the other side of the planet into a firing geometry. The missile, when released, destroyed a vehicle. The path from request to release did not close on the individual the crew believed it had found.

The argument developed here is that the trust failure is remote-kill command and control, not the Hellfire. LrrK’s objects earn their place only where they keep that claim on separate plates: a Campaign for the first-night hunt, a Control Fabric for the CIA–Air Force–CENTCOM seam, a Passport for air vehicle 3034, and a Watch desk that refuses to treat an unpublished rule of engagement as a safe negative. The September reconnaissance-only posture is setup, not a second paper. Later continuity arguments about other presidencies are a sibling file and stay out of this one.

## 2. Reconnaissance Without a Shot

The armed Predator did not appear on 7 October as a finished joint weapon. It arrived as a system that had already seen the target set without a timely way to Act.

Between September and December 2000, with Air Force crews and an interagency analysis team, the Central Intelligence Agency flew an unarmed Predator on fifteen reconnaissance missions over Afghanistan. The Taliban detected the aircraft and launched interceptors against it. On at least one mission the air vehicle imaged a tall man in white robes at Tarnak Farms whose physical and operational signature fit Usama bin Ladin, with a group of about ten people apparently paying their respects. The flights stopped for winter weather. The lesson the Agency drew was not that the sensor had failed. It was that a sighting without a response option was an expensive way to advertise presence. Targets in Afghanistan were hours away from conventional attack even if a policy decision had been made and weapons were positioned. Some officers also argued that continued reconnaissance would undercut a later armed deployment: the Taliban would detect the flights, as it had the year before, and the host country would be unlikely to tolerate extensive operations once that assistance became visible.3

Those sentences are facts about what the opened CIA “Plan” compilation records. They are not, without a dated strike order, facts about a missed shot in 2000. The 2000 flights are the Control Fabric in its pre-weapon form: Sense without Act, and a political-economy argument that more Sense would raise the cost of a later Act.4

The legal and technical unfinished business continued into early September 2001. Weapons tests between 22 May and 7 June had shown excellent missile accuracy and unresolved fusing questions against some targets. The Agency was authorized to deploy weapons-capable aircraft, but for reconnaissance only. The Director of Central Intelligence did not authorize shipment of the missiles, because the host nation had not agreed to flights by weapons-carrying aircraft, and because the fusing questions remained. After 11 September, approval to ship the missiles was granted quickly. Aircraft and missiles reached their overseas location on 16 September. The first mission, on 18 September, flew over Kabul and Kandahar without carrying weapons.5

September is therefore not a pause in the program. It is a stacked constraint. The airframe could carry a missile. The host nation had not cleared the carriage. The fusing question had not been closed. The command-and-control arrangement that could answer a fleeting opportunity without emptying presidential control had been named, in the Agency’s own compilation, as one of the hardest remaining issues. A Passport on the September posture records weapons-capable aircraft, reconnaissance-only missions, missiles present in-country after the 16th, and no host-nation carriage approval until October. Filling the September line with a combat Hellfire would be envelope inheritance.

## 3. Three Permissions on 7 October

The first combat night compressed three permissions that had been negotiated on different clocks.

Host-nation approval for weapons-carrying flights was granted on 7 October. The first armed mission was flown the same day. Operation Enduring Freedom opened the same night. The National Air and Space Museum’s long description of air vehicle 3034 places that airframe on nighttime patrol with two Hellfires near Kandahar, then under Taliban control, and records that the crew located a convoy associated with Mullah Omar, held a firing position, and requested clearance to engage.6 Those are the dated public facts on which a Campaign object can be signed. They do not require a later memoir to establish that 7 October is the first-combat date.

Richard Whittle’s 2011 Mitchell Institute paper, a later reconstruction and labeled as such here, adds operational texture that the October press did not publish. In that account, Army General Tommy R. Franks, commanding Central Command, ordered the first Hellfire strike launched by a Predator in combat on the first night of the Afghan war. The crew was flying the newly armed aircraft from a ground-control station in the parking lot of CIA headquarters at Langley, Virginia. The crew followed Omar and some followers from his residence in Kandahar to a compound of buildings in the countryside.7 That Langley-parking-lot geometry is the split-operations architecture the 2000 “Afghan Eyes” flights had invented, now carrying a weapon release. It is moderate-confidence later history, not a 7 October press line. A Watch desk in October could not have printed it. A 2026 retrospective can record it as reconstruction and must not promote it to a contemporaneous Fact.

Two air vehicles, in the same later account, were already in the air when the war began: the armed Predator flown for the Agency by the Air Force rapid-acquisition office known as Big Safari, and an unarmed Predator launched from Pakistan and flown by a regular Air Combat Command crew.8 October contemporaneous press did not publish a theater inventory. How many air vehicles were over Afghanistan that night remains unpublished in the opened mid-month articles. The unknown is the count, not the existence of an armed first mission.

For LrrK, 7 October is the night Sense, Move, and Act were first joined on an unmanned airframe in this war. The Campaign is the hunt for a named leadership target. The Passport is 3034. The prize is not that a missile came off the rail. The prize is the clearance timeline that decided which aimpoint the missile was allowed to have.

## 4. The Vehicle Instead of the Building

The museum record is specific about the substitute.

Authorization from Central Command was delayed. Omar entered a structure. Concerns about potential incidental casualties and damage led to a decision to strike a nearby Taliban security vehicle in a ploy to flush Omar into the open. The museum calls the event “highly debated ever since,” and it identifies the vehicle shot as the first Hellfire strike from a Predator and the first successful operational use of a precision-guided unmanned-aircraft-delivered weapon.9

Those sentences are facts about what the National Air and Space Museum now publishes on the object. They are not a complete combat record. They do not name the unpublished rule of engagement that converted a building into a vehicle. They do not give an independently confirmed casualty count. Whittle’s later narrative says the missile destroyed a vehicle and killed “a couple of men assumed to be Omar’s bodyguards.”10 That is a 2011 reconstruction. It is not a number that mid-October 2001 press independently confirmed. Exact casualties on the security vehicle remain unknown. The unknown is not a license to treat the shot as bloodless, and it is not a license to invent a body count.

Contemporaneous press handled the same night as rumor before the Pentagon confirmed the class of weapon. The *Recordnet* reprint of an 18 October *Washington Post* report stated that Seymour M. Hersh’s *New Yorker* account — a Predator tracking a convoy carrying Mohammed Omar, a request for an airstrike turned down at Central Command, and the Agency asked to fire a missile just outside the building where the convoy stopped — “could not be confirmed.”11 A same-week *Independent* piece, drawing on the Pentagon’s mid-month reveal, reported that the aircraft that had tracked Omar earlier in the month and had not opened fire on a building where he had taken refuge was an armed Predator; that on that occasion the aircraft was being controlled by CIA operatives and “there was no one who had the authority to order an attack”; and that the sighting was later passed to Central Command at MacDill Air Force Base. The same article said Navy F/A-18s later hit the building after Omar had fled, and that Defense Secretary Donald Rumsfeld was described as “mad as hell.”12 Those are contemporaneous press claims. They corroborate a delayed building shot and a split of authority. They do not, without an official transcript, establish the F/A-18 follow-up or the secretary’s glass-kicking as official record. Hersh remains, in the opened 18 October file, an unconfirmed second-hand account of the delay.

The honest reading is narrower than the later reputation of the night and stronger than a weapons-test press release. Fact: a Predator held a firing position on an Omar-associated convoy and a Hellfire was released at a nearby security vehicle after a delayed clearance on the building. Inference, moderate: the delay was the first public evidence that remote-kill command and control, not missile performance, was the wartime constraint. Judgment: a system that can loiter over a fleeting individual and cannot close its own release path has not completed Sense-Move-Act. Unknown: the rule-of-engagement text, the exact time from request to release, and the number of people killed on the vehicle.

A Kestrel correlation that would have moved the unknown is a dated pairing of the request, the CENTCOM decision, and the rail time. Until that pairing is public, Watch’s job is to refuse a clean “first-kill” headline. The first combat Hellfire is a 7 October event. The first successful remote precision kill of the intended individual is not established in the opened file.

## 5. A Control Fabric With Three Owners

The architecture that produced the delay was not a single cockpit with a slow lawyer. It was a fabric with three institutional owners and a satellite in the middle.

By mid-October, Defense Department officials were willing to describe the split in the press even as the Air Force chief of staff was still telling a congressional breakfast that armed unmanned air vehicles would arrive “in due course.” The *Washington Post* account reprinted on 18 October is the contemporaneous primary. Predators in the Afghanistan conflict were, according to a source familiar with the operation, handled day-to-day by the Central Intelligence Agency because of the hunt for Usama bin Ladin. After the air vehicles took off, control was turned over to Air Force personnel in the United States. If the satellite link was lost, the Agency had backup operators standing by to take over. Photographic images were routinely transmitted to analysts in the United States. The same report said the Agency had taken control of two armed Predators after the Air Force’s Nevada tests and had begun using them in the bin Ladin collection.13

That is a single-point-of-failure design described in plain language. Takeoff, satellite handover, continental flying, Agency backup, and combatant-command release all sit in the same minute when a convoy stops. The *Independent*’s same-week reconstruction of the joint procedure is consistent with that split: intelligence officers initially survey the film from the Predator’s cameras, Air Force personnel control the aircraft in flight, and Agency operators stand by with additional satellite equipment if the link is lost.14 Neither article publishes the written memorandum of agreement. The memorandum is the Lab disposition a 2001 desk should have logged. It remains unpublished in this file.

The satellite path is the kinematic vulnerability that makes the institutional seam expensive. A later reconstruction of Big Safari’s “Remote Split Operations” describes a signal routed to a satellite over Southwest Asia, to a second antenna, to a ground station in Europe, and across the Atlantic on fiber.15 October press did not print that routing. What it did print is enough for a Control Fabric analysis: operators in the United States, thousands of miles from the battlefield, flying an armed aircraft whose lost-link procedure was an Agency backup rather than an automatic hold on weapons. The 31 October Pentagon background briefing, which refused operational comment, did say that a lost link commanded the aircraft to start flying home and that the control lag was “a fraction of a second.”16 A fraction of a second is a piloting delay. It is not a clearance delay. Collapsing the two is how a missile story is written over a command-and-control story.

### Table 1. Public ownership of Sense, Move, and Act, October 2001

| Function | Who the opened record assigns | What remains empty |
| --- | --- | --- |
| Sense (hunt and imagery) | CIA day-to-day ownership of the bin Ladin / leadership hunt; video to U.S. analysts | Unpublished targeting criteria |
| Move (takeoff, satellite flight, lost-link) | Local launch; Air Force personnel in the United States after handover; CIA backup if the link drops | Written CIA–USAF memorandum; theater air-vehicle count |
| Act (weapon release) | CENTCOM authorization; first combat Hellfire from 3034 at a security vehicle after a delayed building shot | Unpublished ROE text; request-to-release times |
| Legal permission to carry | Host-nation approval, 7 October; INF compliance already staffed | Host-nation legal instrument, unpublished |

The table is architecture, not a recap of monthly signals. Each empty cell is a Lab or Watch task. None of them is a safe default. KAT’s job is to keep the Agency hunt, the Air Force stick, and the Tampa release on separate plates. A Campaign that treats “the CIA drone” as one trusted actor has already collapsed the fabric.

The political economy of the split is the reason it existed. The Agency owned the target set and the overseas relationship that made the launch field possible. The Air Force owned the airframe, the crews, and the satellite flying. Central Command owned war-opening fires in its theater. No single office could accept the political cost of a shot that hit the wrong building on the first night of a war, and no single office could accept the operational cost of watching Omar walk inside. The fabric was built to share those costs. Sharing them produced latency. Latency is not an accident in this design. It is the design.

## 6. Mid-October: The Pentagon Confirms the Architecture

By 17–18 October the armed-unmanned fact was no longer a magazine leak that officials could dismiss.

The *Recordnet* reprint is explicit. Defense Department officials said the United States was for the first time flying armed, unmanned aircraft into combat, controlled by operators in the United States thousands of miles from Afghanistan. The RQ-1s had been outfitted with Hellfire anti-tank missiles, “powerful weapons usually carried on helicopters.” A government official said they had fired their missiles several times. The attacks were described as a turning point because the Air Force could now survey and then shoot from lower altitudes without putting pilots at risk, and because the reach of the system created “the real possibility that the United States could someday fly combat missions without having to put large numbers of military personnel on nearby land bases or aircraft carriers.” Eliot Cohen told the paper the war would “give you the revolution in military affairs.”17

The same story records the public-private gap that a Watch desk has to keep on the page. Hersh’s *New Yorker* piece, dated to the 22 October edition, had already described the Omar convoy and the CENTCOM refusal. Pentagon officials at first dismissed the report, saying the Air Force was still experimenting. Publicly, that remained the Air Force position. General John Jumper, asked at a congressional breakfast on Tuesday about putting weapons on drones, said, “We will have armed unmanned air vehicles in due course. We don’t want to push it any faster than it can reliably perform.” The article then states that it “emerged Wednesday” that the Agency had already taken control of two armed Predators.18

That gap is itself evidence about the Control Fabric. The weapon was in combat. The service chief was still speaking as if reliability testing were the constraint. The constraint that had already been demonstrated on 7 October was not whether a Hellfire could come off a composite wing. Nevada had answered that question in February, with the caveat — printed again in October — that the February test was artificial because Hellfire was designed for treetop helicopters and the Predator would fire from around 10,000 feet.19 Mid-month press is the moment the Pentagon confirmed the operators-in-America fact and still would not confirm the Omar delay. Confirmation of the class is not confirmation of the night.

The *Independent*’s parallel story is useful as a second contemporaneous print of the same Pentagon reveal, not as a substitute official record. It reported a joint CIA–Air Force operation, operators thousands of miles away, Hellfires used “several times,” and a refusal to say who or what the targets were.20 How many air vehicles were in theater was not given in the opened *Recordnet* article. That count stays empty.

For product assurance, 18 October is the day the Control Fabric became a public object. A Passport on the mid-month confirmation records armed RQ-1s, U.S.-based operators, Agency day-to-day ownership, Air Force flying after takeoff, Agency lost-link backup, and an official “several times” without a target list. It does not certify the Hersh building-delay narrative. It does not convert Jumper’s “in due course” into a denial that the shots had occurred. The two statements can sit on the same day because they answer different audiences.

## 7. The Treaty Stamp and the Price of Remote Kill

Arming a long-endurance unmanned aircraft raised a treaty question before it raised a combat one: was the result a ground-launched cruise missile by another name?

The 18 October report recorded that some State Department officials had objected to “weaponizing” any drone aircraft as a violation of the 1987 Intermediate-Range Nuclear Forces Treaty, which required elimination of ground-launched cruise missiles with a range of more than 300 miles. Their argument, as the paper summarized it, was that an armed drone was “essentially a recoverable cruise missile.”21 The government’s public answer arrived, on the record, in a 31 October background briefing on unmanned aerial vehicles. A senior defense official stated that prior to the Hellfire demonstration on Predator, the United States government had performed a treaty-compliance review, and that the weaponized Predator was “fully compliant with the Intermediate-range Nuclear Forces — the INF Treaty.”22 Congressional Research Service later cited the same briefing for the same claim.23

Those are facts about what was said on 31 October. They are not the review text. The briefing transcript was not opened here as a signed legal memorandum; it is an on-the-record assertion that a review had been done and that the result was compliance. Later Air Force narrative dates the Compliance Review Group decision to 22 December 2000 and describes engineers waiting through the autumn with wings on sawhorses so that a missile would not be attached to a flyable airframe before the treaty question was resolved.24 That reconstruction explains the pre-combat staff work. It does not replace the 31 October public stamp. A Passport on the legal object records the briefing sentence, records the State Department objection as reported on 18 October, and leaves the review’s reasoning unpublished.

The political economy of the stamp is the same as the political economy of the split. A recoverable, reusable, satellite-flown shooter is valuable because it is not a one-way cruise missile and because it can wait. Waiting is what makes clearance possible. Waiting is also what makes clearance the constraint. A cruise missile is spent when it is launched; its command path closes early. A Predator loitering with two Hellfires keeps the command path open until someone in Tampa, or Langley, or both, accepts the political cost of the aimpoint. The INF finding that the aircraft was not a cruise missile was therefore not a technical curiosity. It was permission to keep a human release in the loop at global distance. The first-night delay is what that permission looked like in combat.

The briefing’s other honesty is useful. Officials would not discuss operational employment. They would discuss the purpose of the spring and summer experiments: “how well can you control a UAV and have it launch weapons, what kind of communications delays, what kind of controls do you have to have in place, and also, what’s the effect on the aircraft.” They had modified the wings for two Hellfires and had launched at static targets up to 15,000 feet. They described the air vehicles as attritable rather than expendable. A Predator system, they said, was four air vehicles plus ground-control and satellite gear; those numbers are not an October theater inventory.25 Remote kill was being bought as a system, not as a one-off missile.

## 8. October’s Incomplete Stack

The October fight is easy to inflate into a full unmanned inventory. The opened record will not carry that inflation.

A 2003 Congressional Research Service report, looking back across the first year of the war, stated that the Department of Defense used three systems in Afghanistan in 2001 — Global Hawk, Pointer, and Predator — against one type, Pioneer, in the 1991 Gulf War and more than ten types in Operation Iraqi Freedom in 2003. Pointer is the hand-launched AeroVironment FQM-151 already in special-operations and Army inventories.26 That three-type list is a 2001 campaign fact in a 2003 inventory. It is not an October order of battle.

Global Hawk’s first combat sorties are a November event. Treating them as an October combat object would collapse a later arrival into the first-night story. Pointer’s October sortie counts were not fetched. The October combat unmanned-aircraft story that the Pentagon was willing to confirm is Predator.

The incomplete stack is the coverage architecture. A medium-altitude armed hunter plus a backpack scout is not a wide-area high-altitude layer and is not a dense tactical mesh. Sense at theater scale was still mostly manned and satellite. Act from an unmanned airframe was a new, scarce, Agency-handled exception. Move, for the exception, was a satellite path with a named lost-link seam. A Campaign that reports “unmanned aircraft in Afghanistan, October 2001” as a single trusted fleet has already done the collapse KAT exists to prevent.

The political economy follows. Because the armed Predator was scarce, every clearance decision carried inventory cost as well as political cost. Because Pointer could not carry a Hellfire, it could not close the Act the Agency wanted on a high-value individual. Because Global Hawk was not yet a combat October object, no high-altitude still-image layer sat in the same minute as 3034’s video. The first-night crew was not short of a missile. It was short of a finished stack and a finished release path.

## 9. What Assurance Objects Could Honestly Record

A retrospective paper is tempted to complete October with what later became known. The useful exercise is the opposite: Passport entries a late-2001 desk could have signed without cheating, and the empty lines it should have left empty.

Air vehicle 3034 is the cleanest object. A Passport records the manufacturer, General Atomics Aeronautical Systems; the Air Force transfer later memorialized by the museum; the January and February 2001 Hellfire tests on this airframe; the two-missile load on 7 October; the Kandahar patrol; the Omar-associated convoy; the delayed CENTCOM clearance; and the security-vehicle aimpoint. It records “first Hellfire from a Predator” as a museum and briefing-class fact. It leaves unpublished the rules of engagement, the request-to-release clock, and the vehicle casualty count. It labels Langley’s parking-lot ground-control station as later reconstruction. It does not certify the night as a successful leadership kill.

The Campaign object is the leadership hunt, not the airframe. It opens on 7 October with host-nation carriage approval and the first armed mission. It records mid-month Pentagon confirmation that Hellfires had been used “several times” without a target list. It does not swallow November’s Global Hawk or a later Yemen shot. Forward indicators a Watch desk would have logged next — and that this paper does not chase — are the next high-value-individual attempt and whether CENTCOM latency repeated, Predator attrition from icing, lost link, or air defense, and whether Pointer or a successor received an urgent operational push. Those are next-month objects. They are not this argument.

The Control Fabric is the three-owner path. A signed fabric diagram would show local launch, satellite handover, Air Force flying from the United States, Agency backup on lost link, Agency ownership of the hunt, and CENTCOM on release. Lab’s missing October disposition is the written CIA–USAF agreement and the lost-link weapons hold, if any. Until those texts are public, “backup operators standing by” is the most honest lost-link line available. It is not a hold. It is not a transfer of release authority. It is a continuity-of-flight procedure sitting next to a weapon.

Kestrel earns its keep on two empty cells: the next dated pairing of a Predator request and a CENTCOM decision after 7 October, and the first official theater inventory that is not a magazine number. Sense, Move, and Act stay separate. Sense was the Agency’s hunt and the airframe’s turret. Move was the satellite path and the handover. Act was a combatant-command permission that arrived in time for a vehicle and not for a building. Product assurance for a system that can do all three from the other side of the world begins by refusing to treat the missile as the proof that the path worked.

Judgment: the trust failure named in the October filing is the correct failure. The Hellfire is a solved, or at least demonstrated, technical object. The clearance is not. Unknowns remain unknowns. They are not safe.

## 10. Conclusion

October 2001 is a first-combat month that civilian speech still files under the missile. The opened record files it under the path.

On the 7th the host nation approved weapons-carrying flights, the war opened, and air vehicle 3034 put two Hellfires into a nighttime patrol near Kandahar. The crew found a convoy associated with Mullah Omar and asked to shoot. Central Command’s authorization was delayed. The substitute aimpoint was a security vehicle. Mid-month the Pentagon confirmed that armed Predators were in combat, flown from the United States, owned day-to-day by the Agency for the bin Ladin hunt, flown after takeoff by the Air Force, and backed up by the Agency if the satellite died. On the 31st a senior defense official put the INF stamp on the record and still would not discuss the shots. Jumper, in the same mid-month window, was still saying “in due course.”

Those are the facts the October public record will support. The inferences that follow are narrower than the month’s later reputation. The Langley parking lot is later history, useful and labeled. The “couple of men assumed to be bodyguards” is later history, not a confirmed count. Pointer was in the 2001 type list and is not the October combat story. Global Hawk is not an October combat object. Unpublished rules of engagement stay empty. Exact vehicle casualties stay empty. Theater air-vehicle count stays empty.

The durable object is not a new munition. It is a Control Fabric that can Sense a fleeting individual from the other side of the world and still lose him to its own release path. Product assurance for systems that sense, move, and act begins by recording that path as the product. The clearance, not the missile, is what 7 October put on the public books. The clearance is what a Watch desk would have had to monitor next.

Endnotes

1. National Air and Space Museum, “General Atomics MQ-1L Predator A,” object record nasm_A20040180000, long description of the 7 October 2001 sortie. https://airandspace.si.edu/collection-objects/general-atomics-mq-1l-predator/nasm_A20040180000 Smithsonian Institution, same object, long description. https://www.si.edu/object/general-atomics-mq-1l-predator:nasm_A20040180000

2. Ibid., summary and long description: first Predator to test-fire Hellfire and first unmanned aircraft to fire Hellfire in combat; delayed CENTCOM authorization and security-vehicle aimpoint.

3. GlobalSecurity.org, “The Plan - Armed Predator,” compilation of the CIA armed-Predator program, including the 2000 reconnaissance series, the 2001 weapons tests, and the early-September reconnaissance-only deployment. https://globalsecurity.org/intell/ops/the-plan-predator.htm

4. Richard Whittle, *Predator’s Big Safari*, Mitchell Paper 7 (Arlington, Va.: Mitchell Institute for Airpower Studies, August 2011), on the September 2000 Tarnak Farms sightings and the absence of a timely cruise-missile response. Later reconstruction, labeled as such. https://www.mitchellaerospacepower.org/app/uploads/2022/08/MP7_Predator_0811.pdf

5. GlobalSecurity.org, “The Plan - Armed Predator”: missiles and aircraft reached the overseas location on 16 September 2001; first mission 18 September without weapons; host-nation approval 7 October; first armed mission the same day.

6. National Air and Space Museum, object record nasm_A20040180000, long description; GlobalSecurity.org, “The Plan - Armed Predator.”

7. Whittle, *Predator’s Big Safari*, opening narrative of 7 October 2001 and the Langley parking-lot ground-control station. Later reconstruction.

8. Ibid., on two Predators airborne as the war opened — one armed and flown for the Agency, one unarmed and flown by Air Combat Command. Later reconstruction. October contemporaneous press did not publish a theater count.

9. National Air and Space Museum, object record nasm_A20040180000, long description.

10. Whittle, *Predator’s Big Safari*. Later reconstruction. Exact casualties are not independently confirmed in the opened mid-October press.

11. *Washington Post*, “U.S. arming unmanned drone aircraft,” 18 October 2001, *Recordnet* reprint. https://eu.recordnet.com/story/news/2001/10/18/u-s-arming-unmanned-drone/50783324007/

12. “Enter the Predator, the unmanned plane sent in to attack the Taliban,” *The Independent*. Contemporaneous mid-October press, second-hand on the Omar building and the reported F/A-18 follow-up. https://www.independent.co.uk/news/world/asia/enter-the-predator-the-unmanned-plane-sent-in-to-attack-the-taliban-9198637.html

13. *Washington Post* / *Recordnet*, 18 October 2001.

14. *The Independent*, mid-October 2001.

15. Whittle, *Predator’s Big Safari*, on Remote Split Operations. Later reconstruction; not printed in the opened 18 October article.

16. U.S. Department of Defense, “Background Briefing on Unmanned Aerial Vehicles,” 31 October 2001, 2:00 p.m. EST, via Federation of American Scientists. https://irp.fas.org/program/collect/uav_103101.html

17. *Washington Post* / *Recordnet*, 18 October 2001.

18. Ibid.

19. Ibid., on the February Nevada live-fire and the altitude caveat.

20. *The Independent*, mid-October 2001.

21. *Washington Post* / *Recordnet*, 18 October 2001.

22. Department of Defense, “Background Briefing on Unmanned Aerial Vehicles,” 31 October 2001.

23. Elizabeth Bone and Christopher Bolkcom, *Unmanned Aerial Vehicles: Background and Issues for Congress*, CRS Report RL31872 (Washington, D.C.: Congressional Research Service, 25 April 2003), citing the 31 October 2001 briefing for the INF-compliance claim and listing Predator, Pointer, and Global Hawk as the three types used in Afghanistan in 2001. https://www.everycrsreport.com/files/20030425_RL31872_089a1bc2a3f353e69de4557e6614bc453e2049b2.html

24. Whittle, *Predator’s Big Safari*, on the 22 December 2000 Compliance Review Group decision. Later reconstruction of pre-combat staff work.

25. Department of Defense, “Background Briefing on Unmanned Aerial Vehicles,” 31 October 2001.

26. Bone and Bolkcom, CRS RL31872, on the 2001 three-type list and on FQM-151 Pointer.

Selected Bibliography

Bone, Elizabeth, and Christopher Bolkcom. *Unmanned Aerial Vehicles: Background and Issues for Congress*. CRS Report RL31872. Washington, D.C.: Congressional Research Service, 25 April 2003. https://www.everycrsreport.com/files/20030425_RL31872_089a1bc2a3f353e69de4557e6614bc453e2049b2.html

GlobalSecurity.org. “The Plan - Armed Predator.” https://globalsecurity.org/intell/ops/the-plan-predator.htm

National Air and Space Museum. “General Atomics MQ-1L Predator A.” Object record nasm_A20040180000. https://airandspace.si.edu/collection-objects/general-atomics-mq-1l-predator/nasm_A20040180000

Smithsonian Institution. “General Atomics MQ-1L Predator A.” Object record nasm_A20040180000. https://www.si.edu/object/general-atomics-mq-1l-predator:nasm_A20040180000

*The Independent*. “Enter the Predator, the unmanned plane sent in to attack the Taliban.” https://www.independent.co.uk/news/world/asia/enter-the-predator-the-unmanned-plane-sent-in-to-attack-the-taliban-9198637.html

U.S. Department of Defense. “Background Briefing on Unmanned Aerial Vehicles.” 31 October 2001. Federation of American Scientists transcript. https://irp.fas.org/program/collect/uav_103101.html

*Washington Post*. “U.S. arming unmanned drone aircraft.” 18 October 2001. *Recordnet* reprint. https://eu.recordnet.com/story/news/2001/10/18/u-s-arming-unmanned-drone/50783324007/

Whittle, Richard. *Predator’s Big Safari*. Mitchell Paper 7. Arlington, Va.: Mitchell Institute for Airpower Studies, August 2011. https://www.mitchellaerospacepower.org/app/uploads/2022/08/MP7_Predator_0811.pdf

Source note. This paper expands the LC-1 October 2001 Kinematic Threat Brief (collection cutoff 31 October 2001; Drive file 1XqrE9KWTo03V0Xp8TuSKNMAcLjOo2Vxe) and does not reprint its five-signal template. It distinguishes the CIA “Plan” compilation on host-nation approval and the first armed mission; the National Air and Space Museum object record for air vehicle 3034; contemporaneous 18 October Pentagon confirmation via the *Washington Post* / *Recordnet* reprint; the 31 October Department of Defense unmanned-aircraft briefing; and later reconstructions (Whittle 2011; the museum long description as a later institutional narrative of a 2001 sortie), which are labeled as such and are not used as October press. Hersh’s *New Yorker* account is cited only as it appears, unconfirmed, in the opened 18 October article. The *Independent* is used as a second contemporaneous print of the Pentagon reveal and is not treated as official record for the reported F/A-18 follow-up. CRS RL31872 is a 2003 inventory of 2001 types and is not an October order of battle; Global Hawk’s first combat sorties are left outside the October file. September 2001 reconnaissance-only posture is setup, not a second paper. Yemen 2002, later presidencies, 2026 policy briefs, Remote ID, and consumer robots are excluded. Analysis is current as of 19 August 2026 as a retrospective on an October 2001 filing and should be revalidated against any subsequently released official pairing of the 7 October request, CENTCOM decision, and rail time, the unpublished CIA–USAF memorandum, or a contemporaneous theater air-vehicle count.