Security research lane

AI security and red-team research

Updated · Class: Reviewed claim · Review: Reviewed by Dread team guidance · Maintainer: Dread Research · Contact [email protected]

AI security is one concrete research/application lane inside DREAD.TECHNOLOGY. The public claim is research and prototype work, not enterprise-platform parity or certified coverage.

Safe public scope

Safe claims today include adversarial and observability research prototypes, eval and artifact workflows, and red-team methodology pages.

Public project story

Janus is a project story about representation-dependent web retrieval. It demonstrates selective delivery from one URL, not downstream agent deception and not production security coverage.

Janus: one URL, two futures

What is not claimed

DREAD.TECHNOLOGY does not currently claim customer deployments, certifications, automated vulnerability coverage, benchmark wins, or parity with established AI security vendors.

Evidence needed next

Stronger public security pages should attach sanitized reports, eval traces, screenshots, reproducible examples, and clearly reviewed artifact pages.

Canonical links