// 0xSEC_RESEARCH · LOCAL_EVIDENCE_LAB
Verify AI-made content with evidence you can check
Inspect C2PA credentials and invisible Unicode steganography locally in your browser. Zero data uploads, verifiable SHA-256 evidence.
01// Read embedded credentials in the browserconst manifest = await c2pa.read(image_file);// Keep presence separate from trustmanifest.found => true// Absence is not an AI or authenticity verdict→ [EVIDENCE]: CREDENTIAL_PRESENTRead whether an image contains a C2PA manifest before making any provenance claim.
// 01_RESEARCH_ADVISORIES
Selected research
Defensive intelligence and vulnerability disclosures across autonomous AI systems and multimodal models.
research.aipwn.orgContinuous Publication
Primary research surfaces
Click to explore threat models and defensive boundaries
Prompt InjectionModel and Context Layer
Studying how untrusted content crosses instruction and memory boundaries.
▼ Expand threat vector and defense
Agent Tool PrivilegesTool and Sandbox Layer
Examining authorization gaps across autonomous tool-call chains.
▼ Expand threat vector and defense
Runtime ExposureExecution Layer
Reviewing exposed local services, weak authentication, and unintended egress paths.
▼ Expand threat vector and defense
Physical and Embodied AISensor and Actuation Layer
Researching how perception failures can cross into physical actions.
▼ Expand threat vector and defense