Die All-in-One Plattform für Medienverifikation.

Pixel-Forensik, Provenienzprüfung und ereignisbasierte Bildanalyse – unterstützt durch erklärbare KI-Agenten.

experten

Context-based media verification.

More than pixel forensics and provenance checks — we provide context- and event-based image analysis, powered by explainable AI agents.
99,5% detection accuracy
<100ms analysis time
experten

Gretchen AI goes beyond manipulated pixels. It understands the context behind them.

Eine integrierte Plattform für Medienverifikation: Authentizitäts-Score, Quellencluster und Kontextanalyse in einem System.
Context Verification &
Sources

Die Kontextanalyse arbeitet unabhängig vom Authentizitäts-Score und durchsucht das Web nach identischen, ähnlichen sowie ereignisbezogenen Bildern.

Gefundene Referenzen werden in Clustern dargestellt. Mit einem Klick sehen Sie Quelle, Datum, Ort und Kontext – und erkennen Übereinstimmungen oder Widersprüche auf einen Blick.

Die Farben der Darstellung entsprechen dem Quellentyp und können individuell angepasst werden.

Deepfake
Detection

Das integrierte Tool erkennt manipulierte oder KI-generierte Bilder und analysiert visuelle sowie digitale Signale.

Ein Authentizitäts-Score schätzt die Wahrscheinlichkeit einer Manipulation ein und dient als Entscheidungsunterstützung.

Die Bewertung sollte stets gemeinsam mit der Kontextanalyse erfolgen.

Unsere Versprechen

Vertrauen
Wir glauben, dass die Offenlegung der algorithmischen Gewissheit (auch in Zweifelsfällen) der Schlüssel zur Vertrauensbildung ist.
Qualität
Wir arbeiten kontinuierlich daran, die Robustheit gegenüber Einflüssen wie Filter, Weichzeichner und Auflösung zu verbessern. Von cheapfake bis deepfake: wir erkennen über 80 verschiedene Arten der Manipulation.
Flexibilität
Durch unsere einzigartige Architektur und vielen Module, haben wir für diverse Anwendungsszenarien eine passende Lösung.

Gretchen AI goes beyond manipulated pixels. It understands the context behind them.

Our Gretchen dashboard is an intuitive and user-friendly all-in-one platform. Upload your files via drag & drop and receive results within seconds.

Context-aware media analysis across multiple scales

We analyze images across multiple scales — from tiny image regions (patches) used to detect local artifacts to the overall image context.

Our model leverages state-of-the-art vision foundation models and benefits from holistic knowledge gained through training on a wide variety of deepfake types.

Fast and clear
Immediate results for fast decisions.
Understandable and explainable
Image, video, and audio analysis.
Continuously improving
Traceable source checks and metadata.

World-class digital forensics

We analyze images across multiple scales — from tiny image regions (patches) used to detect local artifacts to the overall image context.

Our model leverages state-of-the-art vision foundation models and benefits from holistic knowledge gained through training on a wide variety of deepfake types.

Fast and clear
Immediate results for fast decisions.
Understandable and explainable
Image, video, and audio analysis.
Continuously improving
Traceable source checks and metadata.

Gretchen AI goes beyond manipulated pixels. It understands the context behind them.

Our Gretchen dashboard is an intuitive and user-friendly all-in-one platform. Upload your files via drag & drop and receive results within seconds.

Context-aware media analysis across multiple scales

We analyze images across multiple scales — from tiny image regions (patches) used to detect local artifacts to the overall image context.

Our model leverages state-of-the-art vision foundation models and benefits from holistic knowledge gained through training on a wide variety of deepfake types.

Fast and clear
Immediate results for fast decisions.
Understandable and explainable
Image, video, and audio analysis.
Continuously improving
Traceable source checks and metadata.

World-class digital forensics

We analyze images across multiple scales — from tiny image regions (patches) used to detect local artifacts to the overall image context.

Our model leverages state-of-the-art vision foundation models and benefits from holistic knowledge gained through training on a wide variety of deepfake types.

Fast and clear
Immediate results for fast decisions.
Understandable and explainable
Image, video, and audio analysis.
Continuously improving
Traceable source checks and metadata.

A specialized method that goes beyond conventional reverse image search.

It doesn’t just find copies — it identifies the context and time frame of an event. We search for the same underlying event, which may be captured from different camera perspectives of the same situation — significantly increasing analytical insight.

Manage and track all analyses — organized, filterable, and fully traceable.

Organize context and deepfake analyses, filter by status, date, and favorites, and maintain full control over your media verification workflows at all times.

Configure source weighting and reference selection to match your verification strategy

Define which reference images anchor the analysis. Weight sources based on trust level — from established media outlets to unknown domains. This creates a verification model tailored to your organization, risk tolerance, and workflow.

Technical capabilities

Multi-format support
Analysis of single images (JPG, PNG, TIFF) and video containers (MP4, MOV, AVI).
Frame-by-frame inspection
Each video is decomposed into individual frames. Our AI detects manipulations that appear only for milliseconds.
Bulk upload capability
Process up to 100 assets simultaneously.
API functionality
Processing of up to 1,000 assets per upload. Ideal for importing entire press archives or evidence folders. Seamless integration into existing workflows.
Human-in-the-loop
Maintain full control over evaluation and relevance criteria. Customize source classification across categories such as trusted, unknown, or fact-checking.

Flexibel im Einsatz.
Verlässlich im Betrieb.

Robustness. Transparency. Low false positive rates.

🧭
99.5%
Verlässlichkeit
<100ms
Performance
🔒
GDPR compliant
Datenschutz
🌐
24/7
Verfügbarkeit

Klarheit in einer Welt
voller manipulierter Inhalte.

Context Verification &
Sources
Gretchen AI analyzes sources, metadata, and contextual signals to classify content accurately and reliably.
Deepfake
Detection
Our AI detects synthetic media and digital manipulations in images, video, and audio.
Multimediale Verifikation
Text, image, video, and audio — Gretchen AI verifies content across modalities and delivers consistent, traceable results.

Solutions
for multiple industries.

Journalism

EKI-Fälschungen erkennen, auch wenn sie visuell nicht mehr auffallen
Medien verifizieren in Sekunden, direkt im Redaktionsalltag
Verlässliche Quellen & Kontext, statt endloser Gegenrecherche
Fokus zurück auf Einordnung, Bewertung und journalistische Verantwortung

Media Companies & Stock Providers

Echte Bilder von KI-Content trennen, automatisch & skalierbar
Herkunft und Urheberrecht nachweisen, statt nur zu behaupten
Archive monetarisierbar halten, trotz generativer Bilderflut
Vertrauen in menschliche Kreativität sichern, in einer synthetischen Welt

Defense

Deepfakes in sicherheitskritischer Kommunikation erkennen
Identitäten und Befehle verifizieren, bevor sie Wirkung entfalten
Manipulationen auf Pixel- & Signalebene aufdecken
Lagebild und Befehlskette schützen, wo Fehler fatal wären

Media Companies & Stock Providers

Desinformation früh erkennen, bevor sie Wirkung entfaltet
Offizielle Aussagen verifizieren, gegen gezielte Delegitimierung
Öffentliche Personen & Institutionen schützen, vor Deepfakes
Demokratische Stabilität stärken, durch verlässliche Verifikation

Unsere Versprechen

Newsletter

Research-driven updates on media verification, deepfakes, and digital integrity.

New Gretchen AI features — development and verification
Deepfake detection updates & case studies
Signals on emerging media manipulation trends
No spam. Occasional, relevant updates only
Your sign-up could not be processed. Please try again.
Almost done! Please check your inbox and click the confirmation link to complete your subscription.

By subscribing to our newsletter you agree to our privacy policy .