What makes TeraDact's redaction 'permanent' rather than simple data anonymization?

TeraDact permanently removes sensitive data from the underlying file. Simple anonymization, such as removing a name field or masking a Social Security number, often leaves re-identification pathways intact through combinations of quasi-identifiers (age, ZIP code, gender, diagnosis) or overlays and watermarks can be removed. Unlike these less secure methods, content redacted by TeraDact cannot be recovered by adjusting contrast, removing a layer, or inspecting file metadata.

• For documents: underlying text bytes are destroyed, not simply colored over.
• For video and audio: frames and audio segments are reconstructed without the sensitive content.

TeraDact applies AI-driven contextual analysis to identify not just obvious PII fields but indirect identifiers embedded in free text, document metadata, image content, and audio. The result is a more thorough desensitization that reduces re-identification risk even in sophisticated adversarial scenarios.