AWS Security ChangesHomeSearch

AWS wellarchitected: Rewrite ADVSEC05-BP01 from content validation to data protection compliance

Service: wellarchitected · 2026-11-11 · Documentation medium

File: wellarchitected/latest/video-streaming-advertising-lens/advsec05-bp01.md

Summary

The Video Streaming Advertising Lens best practice ADVSEC05-BP01 is retitled from 'Validate and sanitize content before running a campaign' (ad fraud/content validation) to 'Verify that your advertising workload meets applicable data protection obligations'. The body is replaced with compliance-governance guidance: desired outcomes, anti-patterns, risk level (High), six implementation steps (compliance claims catalog, binding claims to data inventory via Amazon Macie discovery, gating material changes, mapping objectives to controls such as TLS for DSP-SSP transport and confidential computing, assembling evidence with CloudTrail records, revalidating claims/exceptions), plus new related best practices (least privilege, secure data collaboration) and related documents (Macie and KMS compliance validation, GDPR regional boundaries whitepaper, IAB Europe TCF). The previous Amazon S3/Rekognition/CloudTrail service list and S3 object-integrity link are removed.

Security assessment

The change is a documentation rewrite of a Well-Architected security best practice, shifting focus from ad-fraud content validation to privacy/regulatory compliance governance. It documents security-relevant practices (least privilege, TLS in transit, confidential computing, Macie sensitive data discovery, CloudTrail evidence, KMS compliance) but references no specific vulnerability, CVE, or incident, so it is security documentation rather than a fix.

Evidence

+# ADVSEC05-BP01 Verify that your advertising workload meets applicable data protection obligations

Diff

diff --git a/wellarchitected/latest/video-streaming-advertising-lens/advsec05-bp01.md b/wellarchitected/latest/video-streaming-advertising-lens/advsec05-bp01.md
index 6fc12a1..ce4ddf0 100644
--- a//wellarchitected/latest/video-streaming-advertising-lens/advsec05-bp01.md
+++ b//wellarchitected/latest/video-streaming-advertising-lens/advsec05-bp01.md
@@ -7 +7 @@
-Implementation guidanceKey AWS servicesResources
+Implementation guidanceResources
@@ -9 +9 @@ Implementation guidanceKey AWS servicesResources
-# ADVSEC05-BP01 Validate and sanitize content before running a campaign
+# ADVSEC05-BP01 Verify that your advertising workload meets applicable data protection obligations
@@ -11 +11,38 @@ Implementation guidanceKey AWS servicesResources
-Content validation is essential to mitigate ad fraud and block unwanted content from reaching ad audience. 
+Advertising platforms process data across many jurisdictions, business relationships, and delivery channels, and each has obligations from law, contracts, and voluntary frameworks. Adherence to obligations requires tracing each obligation to the data it affects, the purpose it governs, the control that enforces it, and the evidence you kept. 
+
+**Desired outcome:**
+
+  * You have approved obligation records that map data types and processing purposes to applicable requirements, covering obligations from laws and regulations, contracts, and voluntary frameworks. 
+
+  * You have controls that identify new or changed personal data flows before they operate without a compliance decision. 
+
+  * You have traceability from each obligation to a control owner, implementation evidence, exception status, and review history. 
+
+
+
+
+**Common anti-patterns:**
+
+  * Treating compliance as a point-in-time audit and allowing controls to drift between reviews. 
+
+  * Presenting encryption or access-control settings as compliance proof without connecting them to a specific obligation. 
+
+  * Applying one retention or consent policy to every advertising dataset, regardless of purpose or jurisdiction. 
+
+  * Reconstructing evidence after an audit request instead of preserving it when a control operates. 
+
+
+
+
+**Benefits of establishing this best practice:**
+
+  * Finds compliance gaps before they affect viewers, partners, or regulatory commitments. 
+
+  * Reduces audit preparation work by maintaining evidence with each control decision. 
+
+  * Gives engineering teams clear requirements without asking them to interpret laws or contracts. 
+
+
+
+
+**Level of risk exposed if this best practice is not established:** High 
@@ -15 +52,11 @@ Content validation is essential to mitigate ad fraud and block unwanted content
-Consider using Amazon S3 which can serve as a secure, scalable storage solution for advertising content. It allows for simple management and distribution of assets. S3 can be configured with strict access controls and encryption to maintain the security of the advertisement files. Additionally, Amazon Rekognition can be utilized to analyze images and videos in advertisements, verifying they meet solution standards and don't contain inappropriate content. This AI-powered service can detect objects, scenes, and activities in visual content. For additional monitoring and auditing, consider using AWS CloudTrail to provide a record of actions taken by users, roles, or AWS services in the ad serving solution, which is essential for security analysis and compliance audits.
+A control can run as designed and still back an invalid compliance claim when the claim names the wrong data, purpose, people, partner, or period. You cannot resolve overlapping requirements with a single rule that applies the strictest requirement. Longer retention satisfies a financial obligation but breaks data minimization. A transfer's terms can determine how a Region treats the data. 
+
+A claim can become invalid even when every control still works. Enriching a dataset makes it more sensitive, aggregating records changes disclosure risk, and a new use can exceed what you interpreted before. Evidence alone does not always corroborate details, especially when different sources cannot be cross-referenced. A point-in-time snapshot proves little once the data or the control changes between observations. 
+
+Independent proof and data minimization are often at odds. Detailed lineage and event records improve traceability, but they can grow into identity-bearing datasets with their own obligations. Independence weakens when one team interprets the requirement, operates the control, accepts the exceptions, and selects the proof. Design your governance so no single team or person does all of these compliance tasks, and weigh evidence fidelity against its cost and exposure. 
+
+### Implementation steps
+
+  1. **Instantiate compliance claims:** Assign each approved claim a unique ID, authority, effective period, scope selector, control objective, evidence assertion, and accountable owners. Store legal and regulatory obligations, contractual commitments, and voluntary industry-framework claims as distinct claim types in one governed catalog. For a claim that depends on TCF, record the TCF policy version your workload implements, its applicable policy scope, an owner, and a review date. 
+
+  2. **Bind claims to the data inventory:** Use [Amazon Macie sensitive data discovery](https://docs.aws.amazon.com/macie/latest/user/discovery-jobs.html) to analyze supported objects in Amazon S3. Use catalog metadata to register other assets, and use workload lineage to trace derived datasets, destinations, and partner transfers. Generate coverage records from these scoped sources. Record unresolved scope rather than inferring approval. 
@@ -17 +64 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-## Key AWS services
+  3. **Gate material changes:** Compare proposed schema fields, partners, Regions, purposes, and destinations with the applicable claim scopes. Hold a changed use until the obligation authority approves a revised claim or the exception authority records bounded acceptance. 
@@ -19 +66 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * Amazon S3 
+  4. **Bind objectives to owned controls:** Map each objective to its data-protection control. Use [ADVSEC02-BP01 Encrypt DSP to SSP communication in transit using TLS](./advsec02-bp01.html) for partner bid transport, [ADVSEC02-BP02 Protect linear distribution feeds and ad signaling in converged delivery](./advsec02-bp02.html) for linear contribution and signaling, [ADVSEC02-BP03 Implement secure data collaboration with least privileged access and privacy controls](./advsec02-bp03.html) for collaboration, or [ADVSEC02-BP04 Implement confidential computing for cross-party measurement without exposing raw viewer data](./advsec02-bp04.html) for record-level confidential computation. Use [AWS Config conformance packs](https://docs.aws.amazon.com/config/latest/developerguide/conformance-packs.html) and workload tests to evaluate the claim without duplicating the control design. 
@@ -21 +68 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * Amazon Rekognition 
+  5. **Assemble the required evidence forms:** Retain design records, control evaluations, [AWS CloudTrail event records](https://docs.aws.amazon.com/awscloudtrail/latest/userguide/cloudtrail-event-reference-record-contents.html), outcome records, approvals, and provenance for the claimed scope and period. 
@@ -23 +70 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * AWS CloudTrail 
+  6. **Revalidate claims and exceptions:** Alert owners when scope, classifications, controls, evidence, or authority changes. Review exceptions before expiration, verify remediation against the original claim, and preserve the closure decision. 
@@ -30 +77,16 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * [Checking object integrity in Amazon S3](https://docs.aws.amazon.com/AmazonS3/latest/userguide/checking-object-integrity.html)
+**Related best practices:**
+
+  * [ADVSEC05-BP02 Design for compliance across both linear and digital data flows](./advsec05-bp02.html)
+
+  * [ADVSEC02-BP03 Implement secure data collaboration with least privileged access and privacy controls](./advsec02-bp03.html)
+
+  * [ADVSEC01-BP04 Implement least privilege access to protect programmatic workloads](./advsec01-bp04.html)
+
+
+
+
+**Related documents:**
+
+  * [Compliance validation for Amazon Macie](https://docs.aws.amazon.com/macie/latest/user/compliance-validation.html)
+
+  * [Compliance validation for AWS KMS](https://docs.aws.amazon.com/kms/latest/developerguide/kms-compliance.html)
@@ -32 +94 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * [Amazon Rekognition](https://aws.amazon.com/rekognition/)
+  * [Defining boundaries for regional services access](https://docs.aws.amazon.com/whitepapers/latest/navigating-gdpr-compliance/defining-boundaries-for-regional-services-access.html)
@@ -34 +96 @@ Consider using Amazon S3 which can serve as a secure, scalable storage solution
-  * [AWS CloudTrail](https://aws.amazon.com/cloudtrail/)
+  * [IAB Europe Transparency and Consent Framework (TCF)](https://iabeurope.eu/transparency-consent-framework/). TCF is a voluntary industry framework. 
@@ -45 +107 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please
-Fraud detection
+Privacy and regulatory adherence
@@ -47 +109 @@ Fraud detection
-Regulatory adherence
+ADVSEC05-BP02 Design for compliance across both linear and digital data flows