Guides
How to think about lookalike comparisons updates for 2027
Lookalike comparisons in 2027 face a new problem: phone photographs are computed, face processing is on by default, and provenance now comes first.
The recent change in face comparison is not how people compare but what a photograph is now.
An ordinary phone picture is no longer a single exposure; software with opinions about faces assembles it as a composite. Comparing two such images compares two pieces of software as much as two people.
What to take away
- A modern phone photograph is computed from several frames, with face-specific processing applied by default.
- Those defaults smooth texture, brighten and even skin, sharpen eyes and often narrow shadows, which moves every face in the same direction.
- Wide phone lenses used at arm's length produce close-range distortion, and portrait modes simulate an optical look that the lens did not have.
- The processing is not recorded anywhere a reader can inspect, so it cannot be subtracted from a comparison.
What the camera now does on its own
Point a phone at a face and the device detects the face before the shutter, exposes for it, stacks several frames to reduce noise, applies tone mapping, and runs a face-aware pass that smooths and sharpens selectively. None of this is optional on most default settings, and none of it is announced.
What the phone does to a face
- Detects the face before shutter
- Exposes for the face
- Stacks frames to cut noise
- Applies tone mapping
- Face-aware smoothing and sharpening
The result is an image that is more attractive and less specific. Fine skin texture is the first casualty, and fine texture is one of the few genuinely individual things in a photograph. Two people photographed on the same model of phone, on default settings, arrive at the comparison having both been moved toward the same house look.
Simulated depth is not depth
Portrait modes blur the background and make the subject look as if shot with a long lens on a large sensor. The blur is computed from a depth estimate, not by the optics.
Real blur vs simulated blur
Real background blur
- Source of blur
- Lens and sensor geometry
- Facial geometry
- Matches the taking lens
- Looks like
- Genuine long-lens shot
Portrait-mode blur
- Source of blur
- Computed depth estimate
- Facial geometry
- Short-lens close-range proportions
- Looks like
- Studio headshot, snapshot face
Real background blur comes from the geometry of the lens and the sensor. The simulated version leaves the face with the proportions of the short lens that took the picture.
A portrait-mode image can look like a studio headshot while carrying the facial geometry of a close-range snapshot. That combination is new, extremely common, and will fool anyone reading the image for cues about how it was made.
The underlying issue is covered in the guide to what camera distance does to a face.
Generated and altered images in ordinary circulation
Alongside the processing that every phone applies, images that are wholly or partly generated now circulate in the same feeds as photographs. Synthesized video and imagery of real people is no longer specialist work, and the visual cues that once flagged it are being closed off steadily.
For anyone comparing faces, provenance is the first question, not a background one. Before asking whether two faces match, ask whether each image is a photograph of a person at all. If it is, ask what was done to it afterward.
Where provenance cannot be established, the comparison has no foundation. Saying so is the honest end of the exercise.
What to recheck, and how often
What to recheck / Why it moves
- Whether an image is a photograph at all
- Generation and editing tools change every few months
- What the capturing device does by default
- Manufacturers change face processing between software versions
- Whether an old image has been reprocessed
- Upscaled and restored versions replace originals in circulation quietly
- Whether a comparison used originals or reposts
- Each repost adds compression and sometimes a crop
- Whether the source image still exists
- Deleted originals leave only the derived versions behind
Nothing on that list is settled by looking harder at the faces, which is the point. It is all provenance work, and provenance work is the part that has become harder rather than easier.
What has not changed
The underlying situation is exactly what it was. Two photographs of two people, made under unknown conditions, do not settle whether the people resemble each other, and no amount of processing on either side improves that. The new tools have made images cleaner, more uniform and less traceable, which pushes in one direction only.
Here the standing position is unchanged too; No pairing of named people is published, and the reasoning is published instead.
And the way a pairing outlives whatever image prompted it is now more relevant rather than less. That is because the prompting image may not survive at all. Scored guessing games inherit the same problem, as the guide to what a quiz actually measures sets out.
Common questions
Does turning off the beautification settings fix this?
It helps, and most devices bury the controls or apply some processing regardless. It also does nothing about images other people took.
Is a raw file a reliable source?
It is much better, because it has not been through the face-aware pass. Almost nothing that circulates is a raw file.
Can I tell whether an image was generated?
Sometimes, from context, metadata or an original that can be traced. Often you cannot, and the correct conclusion in that case is that you cannot.
