Sensitivity and data risk levels

This page describes the data risk and sensitivity levels that Sensitive Data Protectionassigns todata profiles. Tounderstand the data risk levels, it's important to understand the sensitivitylevels first.

Note: When generating data profiles, Sensitive Data Protection scans for only the infoTypesthat you specify in yourinspection template. For example, suppose credit cardnumbers are present in a column. If theCREDIT_CARD_NUMBER infoType isn'tlisted in your inspection template, then the resulting sensitivity and data risklevels for that column don't reflect the presence of credit card numbers.

Sensitivity level

Sensitivity level is an indication of how sensitive the data in a project,table, or file store is. Data is sensitive if it contains detected elements,such as personally identifiable information (PII), financialdata, and credentials.

You can also set the sensitivity of each built-in or custom infoType that youscan for. The sensitivity of each detected infoType affects the resultingsensitivity rating of the profiled resource. For information about how tooverride the sensitivity of a built-in infoType or set the sensitivity of acustom infoType,seeManage infoTypes.

A data profile can have any of the following sensitivity levels:

High
Highly sensitive informationmight be present, including credit card numbers and certain national identifiers.
Moderate
Sensitive information that is not classified as highly sensitive might bepresent. Examples are email addresses and phone numbers, which can be consideredpersonally identifiable. The data might also include freeform text orunstructured data, such as comments.
Low
Sensitive information wasn't detected, and the data doesn't include freeformtext or unstructured data.
Unknown
The data couldn't be scanned successfully. It is uncertain if sensitive data exists.

Sensitivity signals

To calculate sensitivity, Sensitive Data Protection considers the following:

  • Both the default sensitivity of each infoType found along with any useroverrides of the sensitivity.
  • Thelikelihood thathighly sensitive infoTypesare present.
  • Whether the data has an unstructured format and contains mostly freeformtext, like comments.

Data risk level

Data risk level is the risk associated with the data in its current state. Itconsiders the sensitivity level of the data in the resource and the presence ofaccess controls to protect that data.

High
High-sensitivity datamight be present, and there are no access controls to restrict dataexposure. Alternatively, moderate or high-sensitivity data is widely accessible.
Moderate
Moderate-sensitivity datamight be present, and there are no access controls to restrict dataexposure.
Low

The sensitivity level of the data is low. Alternatively, access to the datahas been further restricted, for example, through access controls.

A profiled data asset can also get aLow data risk level if youenabledautomatictaggingand opted to automatically set the data risk of the profiled data assets toLow.

Unknown

The data couldn't be scanned successfully. It is uncertain if sensitive data exists.

Data risk signals

To calculate data risk, Sensitive Data Protection considers the following:

  • The calculated sensitivity level of the data.
  • The presence of access controls that limit access to the data.
  • Whether discovery is configured to set the data risk level toLow whenautomatic tagging is enabled. For more information, seeEnable the automatictagging in the discoveryconfiguration.This option automatically overrides any of the storage-specific formulas.

BigQuery data risk calculation

The following table shows howdata risk signals affectthe resulting data risk level that Sensitive Data Protection assigns to profiledBigQuery resources. TheData risk column shows the resulting datarisk level.

Data sensitivityIs publicColumn policy tag appliedData risk
Low, moderate, or highNoYesLow
Low, moderate, or highYesYesLow
LowNoNoLow
ModerateNoNoModerate
HighNoNoHigh

Cloud SQL data risk calculation

The following table shows howdata risk signals affectthe resulting data risk level that Sensitive Data Protection assigns to profiledCloud SQL resources. TheData risk column shows the resulting data risklevel.

Data sensitivityRequires SSLPublic IP Data risk
LowYesYesLow
LowYesNoLow
LowNoYesLow
LowNoNoLow
ModerateYesYesModerate
ModerateYesNoLow
ModerateNoYesHigh
ModerateNoNoModerate
HighYesYesHigh
HighYesNoModerate
HighNoYesHigh
HighNoNoHigh

File store data risk calculation

The following table shows howdata risk signals affectthe resulting data risk level that Sensitive Data Protection assigns to profiledfile store resources. TheData risk column shows the resulting data risklevel.

Data sensitivityCould contain public filesData risk
LowNoLow
LowInconclusiveresourceVisibilityLow
LowYesLow
ModerateNoLow
ModerateInconclusiveresourceVisibilityModerate
ModerateYesHigh
HighNoModerate
HighInconclusiveresourceVisibilityHigh
HighYesHigh
UnknownNoUnknown
UnknownInconclusiveresourceVisibilityUnknown
UnknownYesUnknown

What's next

  • Learn aboutremediations you can take to reduce data risk and sensitivity.

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2025-12-17 UTC.