Automating responses to authentication requests using unsupervised computer learning techniques
Techniques are disclosed relating to automating authentication decisions for a multi-factor authentication scheme based on computer learning. In disclosed embodiments, a mobile device receives a first request corresponding to a factor in a first multi-factor authentication procedure. Based on user input approving or denying the first request, the mobile device sends a response to the first request and stores values of multiple parameters associated with the first request. The mobile device receives a second request corresponding to a factor in a second multi-factor authentication procedure where the second request is for authentication for a different account than the first request. The mobile device automatically generates an approval response to the second request based on performing a computer learning process on inputs that include values of multiple parameters for the second request and the stored values of the multiple parameters associated with the first request. The approval response is automatically generated and sent without receiving user input to automate the second request.
1. A method comprising:
receiving, by a server system, a login request, wherein the login request corresponds to a first factor in a multi-factor authentication (MFA) procedure;
requesting, by the server system and in response to the login request, a response from a mobile device corresponding to a second factor in the MEA procedure;
receiving, by the server system, a response token from the mobile device, wherein the response token comprises:
an MFA token automatically generated by a machine learning module at the mobile device based on MFA data received from the server system and without receiving input from a user of the mobile device to automate the MFA token; and
one or more context values of the mobile device;
determining, by the server system, that the MEA token is valid and that at least one of the one or more context values complies with a login policy; and
sending, by the server system, an approval response to the login request.
2. The method of claim 1 , wherein at least one of the context values is one of: type of device, proximity to another device, location of device, time of day, current weather, and signal strength.
3. The method of claim 1 , further comprising, prior to receiving the login request:
training, by the server system a plurality of machine learning modules using different context values of a plurality of mobile devices; and
transmitting, by the server system to the mobile device, a trained machine learning module that is unique to the mobile device based on the training being based on a plurality of context values of the mobile device.
4. The method of claim 1 , wherein the mobile device stores at least one value based on the MEA data.
5. The method of claim 4 , wherein the MFA token generated by the machine learning module is generated based on the at least one stored value.
6. The method of claim 1 , wherein the one or more context values that are received at the server system from the mobile device include a frequency of login parameter that indicates how often the user of the mobile device logs into a set of one or more accounts.
7. The method of claim 1 , wherein the one or more context values include a wearable device parameter that indicates whether a wearable device is being worn by the user of the mobile device and whether the wearable device is unlocked.
8. The method of claim 1 , wherein the one or more context values include one or more values that indicate personally identifiable information (PII) that is stored on the mobile device that is not shared with other devices.
9. A non-transitory computer-readable medium having instructions stored thereon that are capable of causing a server computing system to implement operations comprising:
receiving, a login request, wherein the login request corresponds to a first factor in a multi- factor authentication (MFA) procedure;
evaluating data associated with the login request based on a login policy;
requesting, in response to the login request, a response from a mobile device corresponding to a second factor in the MFA procedure, the requesting comprising sending MFA data from the server computing system to the mobile device indicating a required level of security based on the login policy;
receiving a response token from the mobile device, wherein the response token is generated at the mobile device by a machine learning module based on the MEA data and without receiving input from a user of the mobile device to automate the response token; and
sending an approval response based on the received response token.
10. The non-transitory computer-readable medium of claim 9 , wherein the response token is one of:
a null token comprising data which cannot be approved by the server computing system;
an automated token comprising data generated using the machine learning module; and
a user token comprising data based on further requested user input.
11. The non-transitory computer-readable medium of claim 10 , wherein the generating by the machine learning module is based on the MFA data.
12. The non-transitory computer-readable medium of claim 10 , wherein the MFA data comprises one or more context parameters to be used by the machine learning module, wherein the one or more context parameters include environmental parameters for the mobile device.
13. The non-transitory computer-readable medium of claim 9 , wherein a context parameter is one of: type of device, proximity to another device, location of device, time of day, current weather, and signal strength.
14. A method comprising:
receiving, by a server system, a login request, wherein the login request corresponds to a first factor in a multi-factor authentication (MFA) procedure;
evaluating, by the server system, data associated with the login request based on a login policy;
requesting, by the server system and in response to the login request, a response from a mobile device corresponding to a second factor in the MFA procedure, the requesting comprising sending security data from the server system to the mobile device indicating a required level of security based on the login policy;
receiving, by the server system, a response token from the mobile device, wherein the response token is generated at the mobile device by a machine learning module based on the security data and without receiving input from a user of the mobile device to automate the response token; and
sending, by the server system, an approval response based on the received response token.
15. The method of claim 14 , wherein the response token is one of:
a null token comprising data which cannot be approved by the server system;
an automated token comprising data generated using the machine learning module; and
a user token comprising data based on further requested user input.
16. The method of claim 15 , wherein the generating by the machine learning module is based on the security data.
17. The method of claim 16 , wherein the security data comprises one or more context values to be used by the machine learning module, wherein the one or more context values are associated with the mobile device.
18. The method of claim 15 , wherein the machine learning module at the mobile device is unique to the mobile device based on training of the machine learning module including training on a plurality of previous context values of the mobile device.
19. The method of claim 18 , further comprising:
transmitting, by the server system, to a new mobile device utilized by the user of the mobile device, the machine learning module trained on the plurality of previous context values of the mobile device.
20. The method of claim 15 , wherein the mobile device stores at least one value based on the security data, and wherein the generating by the machine learning module is based on the at least one stored value.