AI-generated mood image

How SemantIC works

What happens when SemantIC checks your image.

SemantIC takes an AI-generated image and looks at it along three dimensions: whether it is physically plausible, whether it makes sense in substance, and which clichés it carries. On this page we show you what the tool does, where its criteria come from and – just as importantly – what it deliberately does not do.

01 · The idea behind it

Why good images hide their own flaws.

AI image generators optimize for what reads as “good” at first glance: sharpness, colour, light, composition. They optimize far less for coherence in substance – the logic of the scene, proportions, clichés. Images thus become reliably beautiful, but not reliably correct.

The consequence is the real problem: a convincing image does not invite a closer look. The surface puts the eye at ease, and weaknesses slip through. We call this connection masking – the working hypothesis behind SemantIC: a convincing surface can make critical scrutiny harder. The study behind it supports this assumption; it cannot prove it.

Theoretical basis

This effect is not just an observation; it can be placed scientifically. Recent research on AI detection systems shows that artificial neural networks prioritize visual properties, whereas humans grasp images more through their meaning. That the same holds for image generators is the assumption of this thesis. SemantIC steps into exactly this gap: it looks where the eye is satisfied too quickly.

02 · The process

From image to finding.

SemantIC separates content and aesthetics into two distinct assessment tracks – so that appeal does not gloss over the judgment on substance. Aesthetics are estimated twice: by a vision model and an independent reference model; if the two diverge strongly, the finding says so. Only afterwards are both results brought together.

  1. Input

    Your image

    + optional usage context & original prompt

  2. Two separate checks · parallel

    Check 1 · Content

    Gemini

    Integrity analysis: 3 dimensions (physics · semantics · bias) plus research layer → integrity score.

    Check 2 · Surface

    Claude

    Aesthetic assessment: the visual appeal only, reconciled with an independent reference model → aesthetic score.

    Appeal and Substance are brought together only here – never before.

  3. Merge

    Appeal vs. substance

    Reconciling aesthetics ↔ integrity → masking cues (no single metric).

  4. Finding

    Your result

    Traffic light · 3 dimensions · reading mode · cues.

Model-agnostic – Gemini & Claude are the current standard models, interchangeable. The aesthetic score is the average of Claude and an independent reference model (LAION Aesthetic Predictor); if the two diverge by 20 points or more, the finding says so; if the reference fails, Claude counts alone. If you provide a prompt or context, an image-text comparison (CLIP) also runs – a pure diagnostic signal with no influence on scores and traffic lights. The symbols are stylized placeholders.

02 · b · The guardrails

The guardrails of the assessment.

An image model on its own tends to be led by visual perfection. SemantIC therefore applies guardrails to the assessment that push towards substantiated, traceable judgments.

  1. 01 Tendency

    Gets won over by the beautiful surface.

    01 Guardrail

    Aesthetics are assessed separately – with no influence on the judgment on substance.

  2. 02 Tendency

    Asserts vague impressions (“looks off”).

    02 Guardrail

    Every finding needs visible evidence on the image – otherwise it doesn't count.

  3. 03 Tendency

    Answers freely and associatively.

    03 Guardrail

    Judges only within the fixed research vocabulary (reading modes, drivers, error types).

  4. 04 Tendency

    Ticks off bias categories too hastily.

    04 Guardrail

    Gender and skin tone are only described; bias only on a real pattern (role, power dynamic).

  5. 05 Tendency

    Sticks with the first assessment.

    05 Guardrail

    Deterministic rules re-check the result and correct contradictions.

That makes the assessment more disciplined and more traceable. Where the tool reaches its limits, SemantIC discloses that (section “We know the limits”).

“Masked → clear”: tendency dampened, guardrail sharp. Largely hard-enforced (fixed schema & check rules); the criteria come from 144 coded images.

03 · The three dimensions

What lies behind physics, semantics and bias.

Physics

Physical coherence

Does the image obey the laws of nature? Light, shadow, anatomy, proportions, materials. Such breaks are often small – but they reliably give an image away once you see them.

Semantics

Semantic consistency

Does the scene make sense – and does it fit your context? A technically clean image can still be off in substance: a situation that would not happen that way, or that does not fit your piece.

Bias

Bias and stereotyping

Which roles, poses, clichés does the image show? A beautiful image of a cliché remains a cliché. Important: gender or skin tone are in themselves no finding – only a stereotyped casting of the role, power dynamic or a context break sets it off.

04 · The masking check

When the surface is stronger than the content.

SemantIC assesses separately how good an image looks (its aesthetics) and how well it holds up in substance (its integrity). If appeal clearly exceeds substance, that is the breeding ground for masking – the perfect surface can cover flaws before you notice them. There is deliberately no single masking metric; the quadrant shows the principle, not the result of a check.

The danger is at the top left

If an image sits far to the top left – lots of appeal, little substance – that is the breeding ground for masking. The further a point crosses the diagonal towards the top left, the more strongly aesthetics exceed integrity.

  • Hatched zone – aesthetics exceed integrity.
  • Diagonal – appeal equals substance (parity).
  • Example point – Aesthetics 86 / Integrity 64.

05 · Your input

Why you should frame the image.

Before SemantIC checks, we ask two things: your stance on the image and its intended use. Both only frame the recommendation – they shift neither the score nor the findings.

Stance

What function does the image have?

  • Standard
  • Affirmative
  • Critical
  • Illustrative

The same weakness weighs differently depending on whether an image shows a cliché deliberately or reproduces it unreflectively.

Intended use

What is it meant for?

  • Header image
  • Mood image
  • Symbolic image
  • Illustration
  • Social post
  • Advertising/marketing image
  • Editorial image

From this, SemantIC derives how strict the standard should be.

Selection in the tool – shown here for overview only, not clickable.

06 · A cue, not proof

Why SemantIC proves you nothing.

SemantIC gives no hard verdicts but reasoned cues: each finding shows where you should look more closely – not that an image is “wrong”. Whether it fits your context is not something any tool decides for you; the final judgment stays with you.

What SemantIC is

  • A quality and integrity assessor for AI images
  • A tool that shows you where to look more closely
  • An appraisal before publication – the decision stays with you

What SemantIC is not

  • Not a fake or deepfake detector
  • No “real or fake” verdict about provenance
  • No substitute for your own editorial judgment

07 · Validated limits

We know the limits – and disclose them.

Every limit here has been empirically tested – in documented test series against the human coding, and for anatomy additionally blinded with a second vision model. You should know when you can rely on a finding and when your own eye is needed.

Robust in testing

The red light

In the documented anatomy test run (16 images, 42 runs), no clean image triggered a red finding. If SemantIC shows red, take it seriously – yellow cues, by contrast, can sometimes be off.

Read it right

Green means: nothing found

Green means “nothing found” – not “flawless”. It focuses your visual check; it doesn't replace it.

Known gap

Anatomy: a flag is a check request

The tool finds local hand and finger artefacts more readily than structural body flaws (head, limbs, leg placement) – but it depends on the image: in the tests there were both missed and falsely reported findings, checked across two vision models. An anatomy flag is therefore a check request to you, not a diagnosis.

Framing

Severity is an estimate

Severe, moderate, minor are an estimate by the vision model, not a measurement. Whether a marked point is publication-critical is decided by your own look at the image.

The same tests also surfaced findings that a human eye had missed. The cues complement your visual check – every finding is built to be verifiable with a marking on the image.

AI-generated mood image

08 · Where the criteria come from

Research, not generic model knowledge.

The check logic rests on an empirical basis of its own: a qualitative content analysis of 144 AI-generated images. From it came the recurring error profiles, the reading modes and the visual drivers the tool works with – the practical part of the bachelor's thesis “Visual Bias in the AI-generated Image” (Multimedia Production, FH Graubünden).

The pipeline was developed over documented iterations and tested against the human coding – the validated limits above come from exactly these tests. It is deliberately model-agnostic: criteria, evidence requirements and check rules hang on the research basis, not on an AI model. The models used are interchangeable – the check grid stays.

09 · Privacy

What happens to your image.

Your image is transmitted to external services for analysis and is not stored at SemantIC itself – exactly what goes to whom and how long anything is kept is set out in the privacy policy.

To the privacy policy →

Enough theory. Have an image checked.