Skip to content
Broken UFO Project BROKEN UFO
Wireframe render of a building.
SOURCE PUBLIC EXP-00955 RESOLVED Sighting SECTOR: NON-HUMAN TECHNOLOGY

The 2007 California Drones: Anatomy of a Fraud with a Built-In Language

Anonymous witnesses, impossible objects and a secret dossier with its own alphabet. It lasted a few months and is still circulating without the second half of the story.

OPERATIONAL SUMMARY

Between May and June 2007, photographs of aerial objects with a lattice structure covered in glyphs appeared on the internet, attributed to several anonymous witnesses in northern California, followed by a dossier signed by a certain Isaac that claimed to come from a secret reverse-engineering program called CARET. Independent analyses found convergent indications of computer-generated imagery and identical elements repeated across photographs by supposedly different authors. No witness identified themselves or handed over original files.

1
What is claimedThat several independent witnesses photographed craft of non-human origin and that a program called CARET existed, dedicated to their reverse engineering.
2
What evidence existsThe images and the dossier exist. Their quality and internal coherence are notable, and none of the participants obtained any known financial gain.
3
What doubts remain openNo identified witness, no original file, no verifiable documentary reference, and convergent technical indications of computer synthesis.
8 min
1,690 words
Northern California, United States
V5 · Unverifiable Source: Authorship not established
Start reading
MISSION DEBRIEF EXP-00955

Executive summary of the case file. Everything essential before you dig deeper.

Fact sheet

Code EXP-00955
Source material PUBLIC
Status RESOLVED
Type Sighting
Location Northern California, United States
Opening date 28/07/2026
Last review 28/07/2026

OSINT - Traceability

Verification: V5 · Unverifiable Source: Research Entity: Authorship not established
Opening date 28/07/2026
Status RESOLVED

Some frauds hold up because nobody looks, and some hold up because everybody looks at once. The California drones, in the spring and summer of 2007, belong to the second family. It was fast, coordinated, visually flawless, and it arrived wrapped in a dossier of secret documents with its own alien language built in. It lasted a few months. And in those months it made more noise than cases that have been in the archive for seventy years.

It deserves to be told slowly, because it is a piece of clockwork. This is not a man who sees a light and tells it badly: it is a campaign with structure, with characters, with escalation and with a prepared ending. Whoever built it knew exactly which buttons they were pressing.

And it deserves to be told, besides, because the images are still circulating. You run into them today in compilations, in videos, in filler articles, almost always without the second half of the story. The first half is beautiful. The second is the one that is useful.

PUBLICREF: CARET-2007
Periodspring and summer 2007
Declared locationsnorthern California (Capitola area, Lake Tahoe, Big Basin)
Materialphotographs of lattice-structured objects with glyphs
Documentsthe so-called CARET dossier, attributed to a certain “Isaac”
Authorshipnever publicly established
Statusfraud, by convergent image analysis

What is said to have happened

The sequence begins in May 2007. Photographs signed by a certain Chad, a man from northern California who says he saw and photographed a very strange object, appear on the internet: not a disc, not a light, but a kind of lattice structure, with arms, rings and a belly covered in symbols, floating over a road in broad daylight.

The shape is the first thing that catches the eye, and that is no accident. It breaks with the whole earlier iconographic catalog. It does not look like a flying saucer from the fifties or a triangle from the nineties: it looks like an industrial machine, complicated, with parts that do things. And it is covered in writing.

Over the following weeks more witnesses appear, each with a pseudonym and photos: one at Lake Tahoe, another in the Big Basin area, another still. Different objects, same family resemblance, same glyphs. The accounts are terse and ask for nothing. Nobody sells a book. Nobody gives TV studio interviews. The photos are simply uploaded.

And in June comes the grand finale. An individual who signs as Isaac publishes an extensive dossier: he says he worked in the 1980s on a program called CARET (an English acronym for commercial applications research for extraterrestrial technology), at a facility in northern California, reverse-engineering recovered material. He attaches scanned pages: technical diagrams, device schematics and, above all, an analysis of the language. The same language engraved on the objects in the photographs.

The loop closes by itself. The photos validate the documents and the documents explain the photos. A lone witness can be questioned; a network of independent witnesses who agree on an exotic detail, much less so. And that exotic detail, the writing, is precisely the part that is least likely to be coordinated by chance.

2007-05

Chad

The first photographs of a lattice object covered in symbols appear on the internet, attributed to a witness in northern California.

2007-05

They multiply

Within a few weeks other anonymous witnesses emerge with images of different objects with the same family resemblance and the same glyphs.

2007-06

The CARET dossier

A certain “Isaac” publishes documents he says come from a secret 1980s reverse-engineering program, with an analysis of the same language that appears in the photographs.

2007

The image analysis

Independent researchers detect convergent indications of computer generation: geometry, lighting, depth of field and reuse of elements.

2008

Collapse

The case drops out of serious discussion. Nobody claims authorship; none of the witnesses ever provides verifiable original material.

What can be verified, first

Let us start with what does not exist, which here is almost everything. There is not a single identified witness. All those who contributed images did so under pseudonyms, and none later submitted to an interview with anyone who could check their account. There is no statement before any authority, no complaint, no police report, no air traffic record.

Nor is there any original material properly handed over. A digital photograph carries a technical record inside it (the camera model, the shooting parameters, the date), and that record, together with the unprocessed file, is the first thing requested in any serious verification. In this case none of that was given to anyone who could examine it. Already processed images were uploaded.

And there is no facility, company, contract or budget matching the name of the supposed program. The documents in the dossier carry no verifiable archive reference, no call number, nothing that would allow them to be requested from an agency. They are papers that look like papers.

What does exist is the analysis of the images, and that is where the matter fell apart. Several independent researchers, working separately, found the same kind of indications: geometry too perfect in the structural repetitions, inconsistencies between the lighting of the object and that of the scene, a depth of field that does not match the stated distances, and elements that repeat identically across photographs supposedly taken by different people.

That last point is the one that, for me, settles the matter. If two witnesses who do not know each other photograph different objects in different places and exactly the same glyphs appear in both, with the same typography and the same layout, there are only two explanations. Either the objects come from the same factory, or the images come from the same working file. The second requires a computer; the first, a civilization.

EV-01Reliability: HIGH
Convergent indications of computer generation detected by independent analysts: geometry, lighting, depth of field and reuse of elements across images.
EV-02Reliability: HIGH
No identified witness. All contributors acted under pseudonyms and none handed over original files with their technical data.
EV-03Reliability: HIGH
The CARET dossier lacks archive references, call numbers or any data that would allow it to be requested or checked against a real agency.
EV-04Reliability: LOW
Authorship of the hoax: never publicly established. Profiles compatible with graphic design and three-dimensional modeling have been pointed out, without confirmation.

Why it worked so well

Because nobody asked for anything

The absence of profit is the best alibi there is, and whoever built this knew it. There was no book, no paid lectures, no website with donations. Just uploaded material and silence. That defuses at a stroke the question everyone asks first (who makes money from this?) and leaves the skeptic without his favorite argument.

Because it arrived in layers

A single batch of photographs is examined and discarded. A sequence that grows over weeks, on the other hand, generates a community: people who argue, who compare, who wait for the next installment. By the time the dossier arrived, there was already an audience with the table set and several hypotheses under way. The document did not arrive in a desert: it arrived in a conversation.

Because it came with homework

The language was a brilliant move. Giving people a system of symbols to decipher is handing them a task, and a task turns the spectator into a participant. Months were spent transcribing glyphs, looking for regularities, proposing grammars. All that work, done in good faith by smart people, functioned as publicity and as endorsement: if so many people are studying it, there must be something to it.

WHAT HELD THE CASE UP
Several apparently independent witnesses, with high-quality photographs, a common exotic detail (the writing) and a technical dossier that explained everything. None asked for money or sought personal notoriety.
VS
WHAT SANK IT
No identifiable witness, no original file handed over, no verifiable reference in the documents, convergent indications of synthetic imagery and, above all, identical elements repeated across photographs by supposedly different authors.

Why this case does not die

Because the images are good. As simple and as uncomfortable as that. They survive outside their context: they can be pasted into any compilation with no explanation at all and they still work, because the object is visually memorable. The refutation, by contrast, is a list of technical arguments that does not fit in a photo caption.

And it does not die because authorship was never established. A fraud with a confessed culprit dies on the day of the confession: that is what happened with Maury Island and with so many others. Here there was no confession, and that absence always leaves a crack through which the phrase “it has never been proven false” slips in. There is no need to prove that someone did it: it is enough to show that the images cannot be what they claim to be.

This case file also has a genealogical interest. In 2007, doing this demanded considerable command of three-dimensional modeling and many hours of work. Less than twenty years later, anyone can generate images of that quality in a few minutes. The California drones case is the last great visual fraud of the artisanal world, and that is why it is worth studying: its telltale signs (the light that does not add up, the repeated element, the absence of an original file) are still the ones to look for, only now they have to be looked for much more often.

It is the same school as Gray Barker, who fabricated sources instead of images, and as the Ummo case, which fabricated letterhead documents. The tool changes; the trade does not.

☢
IF YOU ARE GOING TO CITE THESE IMAGES
The photographs of the so-called California drones circulate constantly without context. There is not a single identified witness, no original file was handed over, and independent analyses found convergent indications of computer-generated imagery, including identical elements repeated across photographs attributed to different authors. The CARET dossier contains no verifiable reference. If you use them, use them for what they are: a case study in visual fraud.

How an image like this is examined, step by step

Since the case has aged into a textbook, it is worth writing the textbook down. These are the steps that were followed then and that still hold today, when fabricating an impossible image takes a minute instead of a month.

First, ask for the original. Not the published image, but the file as it came out of the camera, with its embedded technical data: model, lens, aperture, exposure time, sensitivity, date. That block of data does not prove authenticity, since it can be manipulated, but its absence is significant, and a refusal to hand it over even more so. In this case it was never handed over.

Then look at the light. In a real photograph there is a single dominant source, the sun, and everything in the frame obeys it: the direction of the shadows, the color temperature of the lit areas, the bluish fill the sky provides in the shadowed areas. An inserted object usually brings its own lighting, and the discrepancy shows up in the fill sooner than in the main shadows.

Next, the optics. A lens focuses on one plane and progressively defocuses what lies in front of and behind it, and that gradient depends on distance and aperture. If the declared object is a hundred meters away but has the sharpness of something ten meters away, or if its edge does not have the same transition as the real edges of the scene, something does not add up.

Then, noise. Every sensor leaves a grain pattern that varies with sensitivity and that is statistically homogeneous across the image. A rendered element has different noise or none at all, and it is added afterward: the difference is detected by comparing regions.

And finally, what proved decisive here and requires no instrument: look for repetitions. An identical graphic element (the same glyph, with the same proportion and the same layout) present in photographs attributed to people who do not know each other is a shared file. In 2007 that meant one three-dimensional model. Today it would mean the same thing, with less work in between.

What remains are the photos, which is what always remains. An impossible artifact floating over a northern California road, covered in a script that someone invented one afternoon and that a lot of smart people spent months trying to read. Not a small body of work for an anonymous person who never got paid.

OSINT – Source traceability

File completeness33/70
View breakdown
  • +8 Type: investigacion
  • +10 Identified entity
  • +5 Editorial notes
  • +5 Verification level assigned
  • +5 Documented corroboration
Source type Research
Entity / Origin Authorship not established
Verification level V5 · Unverifiable
OSINT editorial notes

No person is named as the author of the hoax: compatible profiles have been pointed out, but there is no public confirmation and this site does not attribute unproven authorship. The image-analysis findings are described by category, without reproducing specific measurements from reports that cannot be consulted.

Sources
[01] Archives of the forums and sites where the images and the dossier were originally published, May-June 2007.
[02] Independent image analyses published in 2007 by researchers in the fields of computer graphics and forensic photography.

FREQUENTLY ASKED QUESTIONS

What were the California drones?
A series of photographs that appeared in 2007 showing aerial objects with a lattice structure covered in symbols, attributed to several anonymous witnesses in northern California.
What is the CARET dossier?
A set of documents published in June 2007 by someone who signed as Isaac and who claimed they came from a secret reverse-engineering program. They contain no verifiable reference.
How was the fraud detected?
By independent image analysis: overly regular geometry, lighting inconsistencies, depth of field incompatible with the stated distances and, above all, identical elements repeated across photographs by supposedly different authors.
Is it known who did it?
No. Authorship was never publicly established, and this site does not attribute unproven authorship. The lack of an identified culprit does not change the nature of the material.
Why do the images keep circulating?
Because they are visually memorable and work out of context, while the refutation is a list of technical arguments that does not fit in a photo caption.
Channel active
Newsletter
SEC:BUFO
Email newsletter

Get a briefing when something moves in the dark.

New case files, declassified documents, recommended reading and signs of anomalous activity. One email when it matters. Zero noise. Zero spam.

No mass mailings
Unsubscribe anytime
OSINT-friendly
Activate subscription
Alternate channel
Follow via RSS
Unit note

We don't sell your email, we don't run campaigns, we don't chase you with cookies. We only send what deserves a place in your archive.

Stable signal No ads One-click unsubscribe