Investigation:
We ran the same tests with SDK v. 18.0.3 and also got 0.7, also we have iOS implementation and scores are different: 0.3 for Fraudulent and 0.7 for Legitimate which means it works on iOS, but we didn’t get 0.9 on manual (human) login. For manual login results are the same.
Unfortunately we are unable to provide support for scoring related questions in this forum, only integration questions. In this case you will need to submit a cloud support request.
Devices: real and simulators
SDK Version: 18.1.1 Google Maven repository Autotest framework: Appium
Steps to reproduce:
Actual result: Scores are always 0.7 after more than 3k activations.
Expected result: 2nd step – score < 0.7; 3rd step – score >= 0.7.
cURL sent accordingly to: Annotating an assessment
Investigation: We ran the same tests with SDK v. 18.0.3 and also got 0.7, also we have iOS implementation and scores are different: 0.3 for Fraudulent and 0.7 for Legitimate which means it works on iOS, but we didn’t get 0.9 on manual (human) login. For manual login results are the same.