Synthetic and Stolen Identity Risk are two aspects of the Identity Verification risk check.
- Synthetic Identity Risk: analyzes the user’s PII for risk they’re using a fabricated or manipulated identity.
- Stolen Identity Risk: analyzes the user’s PII for risk they’re using PII related to a stolen identity.
On a Identity Verification session, these checks will be tagged as either Low, Medium, or High risk. The scoring thresholds for Stolen and Synthetic Identity risk attributes are as follows:
- Synthetic: Low < 60% | Medium < 80% | High >= 80%
- Stolen: Low < 80% | Medium < 89% | High >= 89%
While it's not possible to configure the percent thresholds Plaid uses here, our models have been trained on millions of identities and third party data such as credit header files, phone carrier records, bankruptcies and many more to accurately assess and predict the likelihood of identity theft or manipulated identities. The scoring thresholds above have been fine-tuned based on traffic identified from fraudulent activity and fraud rings across the Plaid network, and have been configured to reduce false negatives while appropriately flagging users with medium and high risk of using a stolen or synthetic identity.