Unverified50% confidenceDecision RuleExact time
If a virtual model requires Zero Data Retention or disallowPromptTraining, these security policies remain enforced even if the request explicitly sets them to false.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedVirtual models can specify which providers to use, configure fallback models, and enforce compliance requirements such as zero data retention (ZDR), so that compliance policies are enforced uniformly at the gateway layer rather than relying on each caller.76% similarUnverifiedGateway-level settings on a virtual model such as fallback, Zero Data Retention and tags switch along with the virtual model, but provider-specific options take effect per provider, so provider options need to be rechecked when switching providers.70% similarUnverifiedEven if a prompt injection successfully alters model behavior, the model cannot execute dangerous operations if network, filesystem, and API access permissions are locked down at the container level.68% similarUnverifiedComputations running in the Ethereum Virtual Machine (EVM) must be deterministic, prohibiting floating-point operations, access to external data, and random number generation.66% similarUnverifiedCAPTCHA blocks automated access through human-versus-machine distinction challenges, but its protective effectiveness is declining against agents driven by multimodal large models.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1109938API
curl https://kongchang.com/api/v1/knowledge/claims/1109938MCP
get_claim(id=1109938)