Formy 3d — agentic threat model

Not certain from the listing — likely uses proprietary or open-source 3D generative models (e.g., diffusion or NeRF-based architectures). Threats include model stealing of proprietary weights and adversarial prompt injection to bypass generation safety limits.

Not certain from the listing — requires large datasets of 3D meshes, textures, and CAD files for training. Threats include training data poisoning (leading to corrupted or biased generations) and intellectual property/copyright infringement claims on training data.

Not certain from the listing — orchestration is likely a simple pipeline converting text/image inputs to 3D formats rather than a complex agentic framework. Threats include insecure handling of input parameters and generation pipeline bypasses.

Not certain from the listing — requires heavy GPU infrastructure for 3D generation and rendering. Threats include resource exhaustion (DoS) due to expensive rendering tasks and potential container escape if user-uploaded assets are processed unsafely.

Not certain from the listing — likely relies on basic user feedback or automated mesh validation. Threats include lack of automated safety filters for inappropriate, offensive, or malicious 3D model generation.

Not certain from the listing — closed-source freemium platform with no explicit security certifications (like SOC2) mentioned. Threats include unauthorized access to user-generated intellectual property and lack of clear data privacy controls.

Not certain from the listing — operates as a standalone platform with no explicit multi-agent or marketplace integrations mentioned. Threats are minimal here, but could involve downstream supply chain risks if integrated into game engines.