It is important for enterprises to enable employees to remotely and securely access official resources. This disclosure describes techniques that enable enterprise identity management providers to review a set of trust-posture signals from a device to enable secure sign-on and access to enterprise resources. A cryptographic nonce is generated and shared with enterprise ecosystem partners, such that they can confirm that the operating system is working as intended. A large dataset is created from snapshots of device trust-posture signals and results of access requests. A large language model is trained on the large dataset. An application programming interface (API) that utilizes the trained large language model is implemented and provided to partners (e.g., app developers) to improve the security of access to enterprise resources from remote devices. Per the techniques, enterprise identity management providers can be assured in a zero-trust manner that the signals have not been tampered with.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.