The General Intelligence Company — agentic threat model

Not certain from the listing — The specific foundation models powering these digital workers are not disclosed, leaving potential exposure to standard LLM vulnerabilities such as prompt injection, model reprogramming, or misaligned outputs.

Not certain from the listing — The data operations, vector stores, and RAG architectures used to ingest business data are unspecified, presenting risks of data exfiltration or knowledge-base poisoning if left unsecured.

Not certain from the listing — The orchestration framework, memory management, and tool-calling mechanisms for business automation are not described, raising concerns regarding insecure tool integration and unauthorized action execution.

Not certain from the listing — The deployment infrastructure, sandboxing controls, and network isolation measures for hosting these digital workers are unknown, which could lead to host compromise or lateral movement if compromised.

Not certain from the listing — There is no mention of evaluation frameworks, real-time monitoring, logging, or guardrails to detect drift, anomalies, or malicious inputs in production environments.

Not certain from the listing — The listing does not detail any identity management, access controls, policy enforcement, or regulatory compliance alignments (such as SOC2 or ISO 27001) for the automated workflows.

Not certain from the listing — While 'Digital Workers' implies potential multi-agent coordination or integration into broader enterprise ecosystems, the specific interaction protocols and trust boundaries between agents are undefined.