Provision record
AWS Bedrock · AWS Service Terms · View original document ↗

Prohibition on Circumventing Model Safety Features

Medium severity Medium confidence Explicit document language Common · 281 of 352 platforms
Stay ahead of the changes
Track AWS Bedrock and get the diff the day its terms change.
Share 𝕏 Share in Share 🔒 PDF
Document Record

What it is

The terms prohibit customers from attempting to disable, bypass, or otherwise circumvent safety filters and content moderation mechanisms built into Amazon Bedrock or the foundation models accessible through the service. Violations of this restriction may constitute a breach of the Service Terms.

This analysis describes what AWS Bedrock's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology

ConductAtlas Analysis

Why it matters (compliance & governance perspective)

This provision establishes an enforceable restriction on prompt engineering or technical approaches designed to override model guardrails, which is operationally significant for red-teaming, security research, and adversarial testing use cases that may require evaluation of model safety boundaries.

Interpretive note: The precise scope of what constitutes circumvention versus legitimate prompt engineering or model evaluation is not defined in the provision, creating operational ambiguity for security research and adversarial testing use cases.

Recent Activity

This document changed recently

Medium Jul 30, 2026

The updated terms now explicitly state that AWS IoT SiteWise Scenario Discovery is not designed for real-time vehicle control and cannot be used as the sole basis for determining vehicle safety or regulatory compliance. Organizations deploying this service must implement independent human monitoring and safety validation before using its outputs to support vehicle system decisions. The terms make clear that AWS assumes no responsibility for uses that violate these constraints.

View change record →
Medium Jul 18, 2026

The updated terms establish new restrictions on how AWS Capacity Reservations may be used. Specifically, customers purchasing On-Demand Capacity Reservations can no longer resell them to other parties, and AWS reserves the right to cancel the purchase or terminate running instances if the company suspects resale activity. For Capacity Blocks for ML, the grace period before instance termination increased from 30 minutes to 60 minutes for UltraServer instance types, allowing slightly more time to complete workloads. The Amazon Sidewalk qualification program was renamed and simplified, but the underlying security and operational requirements remain in effect.

View change record →
Medium Jun 30, 2026

The updated terms establish a formal framework for AWS Bedrock's free exploration services, clarifying the operational boundaries and responsibilities. AWS reserves the right to discontinue these services at any time without prior notice, meaning customers cannot rely on their continuation for production planning. Customers are solely responsible for testing, deploying, and maintaining any code, documents, or AI solutions AWS provides, including determining whether those solutions comply with applicable law. AWS retains the right to develop competing products based on content it creates during these engagements, though this does not override existing non-disclosure agreements. Customers are prohibited from requiring AWS personnel to sign additional terms as a condition of receiving free services, and any such documentation signed by AWS personnel is void.

View change record →

Clause Stability Mostly Stable

1
Change
3
Months Monitored
May 20, 2026
First Seen
May 22, 2026
Last Seen
This clause type exists across 4245 other provisions on other platforms.
This clause has changed once in 3 months of monitoring.

Change history

removed Jul 11, 2026

The removal of this prohibition on bypassing AI safety mechanisms eliminates an important guardrail against adversarial use of AI services.

View full change record →
modified May 23, 2026

Language simplified from 'attempt to disable, circumvent, or otherwise undermine safety mechanisms, content filters, or use policies' to 'disable or circumvent any safety filters or mechanisms,' removing redundancy and focusing on core prohibited actions.

View full change record →

Consumer impact (what this means for users)

Under this clause, customers are prohibited from using techniques designed to bypass or disable safety controls in Bedrock models, which applies to all customer accounts regardless of use case or stated purpose. Security researchers and organizations conducting AI red-teaming should assess whether their activities require separate authorization from AWS.

How other platforms handle this

Atlassian Medium

interfere with or circumvent Product usage limits or Scope of Use restrictions

Tinder Medium

You may not display any personal contact, banking, or peer-to-peer payment information, whether in relation to you or any other person (for example, names, home addresses or postcodes, telephone numbers, email addresses, URLs, credit/debit card...)

ActiveCampaign Medium

Bypass or ignore instructions contained in our robots.txt file that controls automated access to portions of our Services;

See all platforms with this clause type →
▸ View Original Clause Language DOCUMENT RECORD
"
You may not use Amazon Bedrock to disable or circumvent any safety filters or mechanisms built into Amazon Bedrock or the models available in Amazon Bedrock.

Excerpt from AWS Bedrock's AWS Service Terms

ConductAtlas Analysis

Institutional analysis (regulatory & governance intelligence)

(1) REGULATORY LANDSCAPE: This provision engages emerging AI safety regulatory frameworks including the EU AI Act's requirements for maintaining safety measures in AI systems and the NIST AI Risk Management Framework's guidelines on adversarial robustness.

Insight

Unlock the full institutional analysis

Enforcement risk, jurisdiction flags, contract triggers, and due diligence action items.

Applicable agencies

  • Federal Trade Commission (ftc)
    Oversees unfair or deceptive business practices and can investigate companies that mislead consumers about data collection, sharing, or use.
    Who can file: Anyone affected by the company's practices (US or international)
    What you need: Your account details, a timeline of relevant events, and a description of the specific issue
    What to expect: Complaints inform FTC enforcement priorities and investigations but do not result in individual resolution or compensation
    File a complaint →

Applicable regulations

CFAA
United States Federal

Provision details

Document information
Document
AWS Service Terms
Entity
AWS Bedrock
Document last updated
May 5, 2026
Tracking information
First tracked
May 20, 2026
Last verified
May 20, 2026
Record ID
CA-P-012382
Document ID
CA-D-00648
Evidence Provenance
Source URL
Wayback Machine
Content hash (SHA-256)
0169685ed8ddb3a5676650839f1053019501dd43ed3f5df355022e3aff0aac80
Analysis generated
May 20, 2026 20:43 UTC
Methodology
Evidence
✓ Snapshot stored   ✓ Hash verified
Citation Record
Entity: AWS Bedrock
Document: AWS Service Terms
Record ID: CA-P-012382
Captured: 2026-05-20 20:43:05 UTC
SHA-256: 0169685ed8ddb3a5…
URL: https://conductatlas.com/platform/aws-bedrock/aws-service-terms/provision/CA-P-012382/prohibition-on-circumventing-model-safety-features/
Accessed: Aug. 12, 2026
Permanent archival reference. Stable identifier suitable for legal filings, compliance documentation, and research citation.
Classification
Severity
Medium
Categories

Other risks in this policy

Get the research letter

Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.

Frequently Asked Questions

What does AWS Bedrock's Prohibition on Circumventing Model Safety Features clause do?

This provision establishes an enforceable restriction on prompt engineering or technical approaches designed to override model guardrails, which is operationally significant for red-teaming, security research, and adversarial testing use cases that may require evaluation of model safety boundaries.

How does this clause affect you?

Under this clause, customers are prohibited from using techniques designed to bypass or disable safety controls in Bedrock models, which applies to all customer accounts regardless of use case or stated purpose. Security researchers and organizations conducting AI red-teaming should assess whether their activities require separate authorization from AWS.

How many platforms have this type of clause?

ConductAtlas has identified this type of provision across 281 platforms. See the full comparison.

Is ConductAtlas affiliated with AWS Bedrock?

No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by AWS Bedrock.