We make films, photos,
and voice with AI.
We also tell you when someone else's are fake.
Abundera Studio is a production house with a forensics desk. Every frame we ship is signed with C2PA Content Credentials. Every frame we're sent gets checked.
Two arms, one studio.
We create with AI where it earns its place, and we run a lab that catches what shouldn't be trusted.
Production
Films, stills, voice, score, and copy — produced under consent, watermarked at export, verifiable by anyone with a browser. SAG-AFTRA-aligned likeness contracts. C2PA Content Credentials on every signed asset. SynthID where the model supports it. Per-asset model provenance disclosed.
Forensics
Deepfake video, voice clones, manipulated images, and AI-written text. We run cross-modal detection, signal-of-origin checks (C2PA, SynthID), and a manual review pass on contested cases. 24-hour turnaround on standard files; same-day on urgent.
What we cover
Video
Make: ad spots, social shorts, product walkthroughs, AI-supported edits with on-screen disclosure where required.
Verify: face-swap detection, lip-sync mismatch, frame-level model artifacts, re-encode lineage.
Voice + audio
Make: voiceover, narration, podcast intros, score and SFX, with consent records for every cloned voice.
Verify: voice-clone detection, splice/edit boundary analysis, SynthID audio detection where present.
Images
Make: brand stills, product photography, illustration, with C2PA signing baked into export.
Verify: AI-generation detection, splice/edit forensics, EXIF + provenance trail audit.
Text
Make: copy, scripts, longform, with human-edit attribution recorded under EU AI Act Article 50.
Verify: AI-text detection across major models, tone/style fingerprinting, plagiarism + paraphrase scan.
How we work
Brief
Tell us what you need: a campaign, a single asset, or a forensic check on a file. We scope, we quote, we book.
Produce or examine
Production runs on a consent-first pipeline; forensics runs through cross-modal detectors plus a human review pass. Every step is logged.
Deliver, signed
Deliverables ship with C2PA Content Credentials, watermarks where supported, and a one-page provenance brief. Forensic reports include the verdict, the evidence, and the limitations.
The standards we work to
No marketing claims — the actual list of frameworks, tools, and laws our pipeline maps to.
C2PA Content Credentials
Every Studio asset is signed at export. The credential travels with the file and verifies in any browser via Content Authenticity Initiative tooling.
SynthID + watermarking
Where Google's SynthID or comparable signal is available (Veo, Imagen, Lyria, certain Sora outputs), we preserve and surface it.
EU AI Act Article 50
Disclosure obligations enter force August 2026. Our deliverables include the markings and disclosures the deployer needs to comply.
SAG-AFTRA AI provisions
Performers are engaged under contracts that match the union's transparency, consent, compensation, and control framework, including the right to suspend consent.
California SB 942
Effective Jan 1, 2026, large GenAI providers must publish a free detection tool. Studio's forensics arm is built for that volume of incoming files, not just internal QA.
Detection stack
Multi-vendor cross-checked — we don't trust a single detector. Combined with provenance signals and a manual review for contested cases.
Have a file you need verified?
For newsrooms
Same-day turnaround on time-sensitive deepfake checks. We deliver a verdict, a confidence interval, and the evidence that supports both.
For brands and legal
Verified file → sworn-statement format suitable for cease-and-desist, takedown, or litigation prep. Chain-of-custody log included.
For platforms
Bulk pipelines for trust + safety teams. We integrate behind your moderation queue, not in front of your users.
Production slots and forensic intake open soon.
We're booking initial commission and verification work now. Leave your email and we'll be in touch with a specific slot, not a marketing list.