Viewer-funded independent journalism Sign in Sponsor the investigation
Scenario 3

Scenario 3: the gun drone

Pseudonymous troll “John Cullen’s” account: Elbit Systems Bird of Prey – a rifle-carrying drone over the Computer Science building, 302 m (330 yd) from Charlie.

3D model view: Scenario 3: the gun drone
Drag to move · Scroll to zoom · Shift + drag to rotateOpen full screen

The account

John Cullen says the shot was fired by a rifle-carrying drone, which he identifies as an Elbit Systems Bird of Prey. He has given two versions of where it was.

To Jimmy Dore he repeatedly named the Science Building. To Bret Weinstein he named the Computer Science building, 330 to 350 yards from Charlie at an angle of about 10 degrees. Those are different buildings on opposite sides of the courtyard. The model tests both, as Scenario 3a and Scenario 3b in the viewer’s Views menu.

What the model shows

Scenario 3a, the Science Building. The drone is placed over its roof, north-west of the courtyard, 218 m (239 yd) from Charlie and 60 m above him. The line from there stops at the concourse building, 24 m (79 ft) short of him. No angle on that bearing gives a clear line.

Scenario 3b, the Computer Science building. The drone is placed where the second version puts it: 302 m (330 yd) from Charlie at 10 degrees, which is 52 m (172 ft) above him and about 16 m above the building’s roof. That line is clear. It passes over the edge of the Losee Center roof and under the edge of the tent.

The opening is narrow. On that bearing the line is clear only between about 9.4 and 13.4 degrees: lower and the Losee Center blocks it, higher and the tent does. Like the Scenario 1 line, it passes about 0.6 m above Hunter Kozak’s head.

Seen from the amphitheater, a drone in the 3b position would sit just above the Losee Center roofline, about one degree to the right of the Scenario 1 roof position. In the viewer a red ring marks the spot.

Scenario 3a: the view from over the Science Building. The line to Charlie is blocked.
Scenario 3a: the view from over the Science Building. The line to Charlie is blocked.
Scenario 3b from the roof deck where Adam Bartholomew stood, turned to face the drone position. The ring marks where the drone would be, just above the Losee Center roof.
Scenario 3b from the roof deck where Adam Bartholomew stood, turned to face the drone position. The ring marks where the drone would be, just above the Losee Center roof.

What the video shows

Adam Bartholomew was recording on the roof deck with a MEVO camera. During an interview there his camera stood still in two positions, and both had the 3b drone position in frame: from 137 seconds before the shot until 25 seconds before it.

The direction of the camera was worked out from mountain peaks that also appear in the frame already matched to the model at the time of the shot. That puts the drone position in his picture to within about 1.5 degrees. A drone of the size modelled would be about 6 pixels wide there.

All 3,265 frames from that period were cut out around the spot and combined in 8-second groups, which removes noise and makes anything that stays in place stand out. Every group shows clear sky. Only cloud wisps, the mountain behind and one person’s hair enter the cut-out.

For the last 25 seconds and the shot itself, his camera had turned to face the tent and the spot was 26 degrees outside his frame. So his video rules out a drone waiting at that position in the two minutes before the shot. It does not cover a drone that arrived in the final 25 seconds.

A small white speck appears elsewhere in the same sky, about 21 degrees to the right of the drone position and 4 degrees higher. It forms out of a group of faint wisps and drifts north at the same rate they do. It is not hovering. It is most likely a small cloud fragment; its distance cannot be measured from one camera.

The Not Sameer Massey and Cub Keller phone positions could not have seen the 3b position: it was behind one and well outside the frame of the other, and hidden by a building or a tree from both.

Adam Bartholomew’s camera 47 seconds before the shot. Red ring: the Scenario 3b drone position. Yellow box: the white speck.
Adam Bartholomew’s camera 47 seconds before the shot. Red ring: the Scenario 3b drone position. Yellow box: the white speck.
The sky around the drone position, frames combined in 8-second groups from 137 to 25 seconds before the shot. The ring is one degree in radius.
The sky around the drone position, frames combined in 8-second groups from 137 to 25 seconds before the shot. The ring is one degree in radius.
The white speck over 40 seconds. It forms from a wisp and drifts with the wisps around it.
The white speck over 40 seconds. It forms from a wisp and drifts with the wisps around it.

What the testimony says

Court testimony on YouTube. Starts at the questions about the round, 33 minutes in.
Agent Brian Davis, Utah State Bureau of Investigation, at the preliminary hearing. The video starts where he is asked about the round found on the roof.

A round of ammunition was found on the roof of the Computer Science building on 10 September 2025. Agent Brian Davis of the Utah State Bureau of Investigation, one of the two case agents, testified about it at the preliminary hearing.

It was a live round, not a spent shell casing. Davis said it was intact and had not been fired: the primer was not dented and the bullet was still seated in the case. He called it “a fully functional bullet”. A rifle that has fired leaves an empty case behind. This was a complete cartridge, so no shot produced it.

It was a .223, not the .30-06 of the prosecution account. It lay on the far east side of the roof, the end furthest from the amphitheater. Davis said that according to the agent who found and collected it, there was no line of sight from there to the tent. The model agrees: from the east end of that roof, the building itself and then the Losee Center are in the way.

Agent David Hull, questioned the day before, said an unfired round found on another building was accounted for as a cartridge ejected by an officer who had cycled his rifle. He did not recall which building or what type of round it was.

To be fair to the account, the model does find a clear line to Charlie from the height it gives above that building, as set out above. What the account does not explain is the round. A rifle that fires throws out an empty shell casing. A live cartridge leaves a rifle only when the action is cycled without the round going off: when the weapon is unloaded or cleared by hand, or after a misfire. A drone built to fire a rifle from the air has no reason to drop an unfired round on the far side of the roof below it and it would not even seem to have the capability to do that. An officer unloading a patrol rifle on that roof would have that ability but this unfired round raises more questions than it answers.

The round was also not found under the drone. The building is about 83 m long. At the range and angle the account gives, the drone sits over the middle of it, and the open roof at the east end is 28 to 55 m from the point beneath the drone, on the side away from the amphitheater. At the account’s longer range of 350 yards that becomes 10 to 39 m. By a rough estimate, a cartridge thrown from a rifle 16 m above the roof would land within about 10 m.

The exact spot has not been made public. In the model, an orange outline marks the open roof at the east end of the building, about its last 10 to 20 m, as the best reading of “the far east side”. The pin is an estimate inside that area, not a surveyed point. It sits in the middle of the open roof there, clear of the rooftop equipment and at least 6 m in from any edge, on the reasoning that an officer clearing a rifle would be standing on the open roof, not at the parapet. It is about 40 m from the point under the drone, or about 25 m at 350 yards. If the words meant the long edge of the roof that faces south-east instead, the drone is about 24 m from that edge.

So the one physical item recovered from that roof is an unfired cartridge. It shows that a .223 firearm was handled there at some point. It is not evidence that a shot was fired from the building or from above it, and the testimony reviewed here describes no spent case, no drone and no shot from that direction.

The Computer Science building from above. Left ring: the Scenario 3b drone, 16 m above the roof. Orange outline: the open roof at the east end, the area the testimony points to. Right ring: the model’s estimate within it, set in from the roof edges.
The Computer Science building from above. Left ring: the Scenario 3b drone, 16 m above the roof. Orange outline: the open roof at the east end, the area the testimony points to. Right ring: the model’s estimate within it, set in from the roof edges.

What has not been tested

The 25 seconds before the shot, and the shot itself, at the 3b position. A video that shows the Losee Center roofline during the shot would cover it.

Positions other than the two the account names.

Whether a drone that small would show at 302 m in this camera. The combined frames are clean enough that it should, but that has not been checked against a real drone at a known distance.

The Computer Science building lies outside the surveyed area. It is built from its map outline and aerial imagery, with its height estimated from the number of storeys.

Whether such a drone could hold steady enough for that shot, and whether it would have been heard. Neither is modelled.

Scenarios are interactive studies for investigation, not findings.

Related coverage

Part of The Charlie Kirk Assassination: Interactive Forensic Investigation Tool, a measured 3D model of the Utah Valley University amphitheater as it stood on 10 September 2025.

Account
John Cullen, in two versions
3a: Science Building
218 m (239 yd). Line blocked, 24 m short of Charlie
3b: Computer Science building
302 m (330 yd) at 10 degrees. Line clear
Clear angles for 3b
About 9.4 to 13.4 degrees
Height of the 3b drone
52 m (172 ft) above Charlie
Video check
Nothing at the 3b position from 137 to 25 seconds before the shot
Round on the roof
A live, unfired .223 cartridge, not a spent casing; far east side of the Computer Science roof (testimony of Agent Brian Davis)
Round to drone
About 40 m from the point under the drone (best estimate)

Viewer-funded journalism

This investigation is paid for by the people who watch it.