Anthropic · Anthropic Responsible Scaling Policy · View original document ↗

Self-Reported RSP Compliance Gaps

Medium severity High confidence Explicitdocumentlanguage Unique · 0 of 352 platforms
Get alerted the next time Anthropic changes these terms. Get same-day alerts →
Share 𝕏 Share in Share 🔒 PDF
Recent governance activity Anthropic recorded 3 documented changes in the last 30 days.
Get same-day alerts →
Monitor governance changes for Anthropic Monitor emails you the same day this changes. The archive stays free.
Get same-day alerts →

Get the weekly 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. No account.

Document Record

What it is

Anthropic discloses that in implementing the prior RSP, the company identified several instances of non-compliance with its own requirements, including delayed evaluation timelines, missing elicitation techniques, and evaluations not explicitly designed to meet stated scaling buffer requirements.

This analysis describes what Anthropic'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 discloses that Anthropic's adherence to its own stated RSP requirements has not been complete, with specific procedural gaps identified in evaluation timelines, methodology, and buffer assessment; the company characterizes these gaps as minimal risk and uses them as a basis for policy revisions.

Consumer impact (what this means for users)

The document discloses that evaluations conducted under the prior RSP version did not fully meet all stated requirements in at least four specific procedural areas, and that the updated policy incorporates revisions intended to address those gaps by adding flexibility and improving compliance tracking.

Cross-platform context

See how other platforms handle Self-Reported RSP Compliance Gaps and similar clauses.

Compare across platforms →

Monitoring

Anthropic has changed this document before.

Receive same-day alerts, structured change summaries, and monitoring for up to 25 platforms.

Get Monitor Or create a free account →
▸ View Original Clause Language DOCUMENT RECORD
"
As part of this, we reviewed how well we adhered to the framework and identified a small number of instances where we fell short of meeting the full letter of its requirements. These areas were: Our most recent evaluations were completed 3 days later than the 3-month interval... Some of our evaluations lacked some basic elicitation techniques such as best-of-N or chain-of-thought prompting... Our evaluations in one domain were not explicitly designed to establish the 6x scaling buffer mentioned in the previous policy.

Excerpt from Anthropic's Responsible Scaling Policy

ConductAtlas Analysis

Institutional analysis (regulatory & governance intelligence)

1) REGULATORY LANDSCAPE: Self-disclosure of compliance gaps with voluntary safety commitments is relevant to FTC assessment of whether public representations about AI safety practices constitute unfair or deceptive claims. In the EU, where AI providers may be required to maintain technical documentation demonstrating conformity with AI Act requirements, gaps in evaluation methodology could affect conformity assessment outcomes. 2) GOVERNANCE EXPOSURE: Medium. The self-reported compliance gaps are limited to procedural matters and are characterized as posing minimal risk in the document. However, the disclosure that evaluations lacked basic elicitation techniques and that scaling buffer requirements were not formally met may be material to organizations relying on Anthropic's evaluation outputs for their own risk assessments. 3) JURISDICTION FLAGS: Organizations in EU/EEA jurisdictions using Anthropic's models for high-risk applications should evaluate whether the disclosed evaluation gaps affect the reliability of safety documentation they have relied upon. US federal procurement contexts may similarly require assessment of whether these gaps affect vendor qualification. 4) CONTRACT AND VENDOR IMPLICATIONS: Vendor assessment frameworks that incorporate Anthropic's RSP compliance as a criterion should be updated to account for the disclosed gaps and the policy revisions made in response. Due diligence questionnaires for Anthropic should specifically address evaluation methodology and timeline adherence going forward. 5) COMPLIANCE CONSIDERATIONS: Organizations that have cited Anthropic's RSP compliance in their own regulatory submissions or risk assessments should review whether the disclosed gaps require disclosure updates or reassessment. The revised RSP's introduction of a 6-month evaluation interval and improved compliance tracking mechanisms should be noted in updated vendor assessments.

Full institutional analysis

Regulatory citations, enforcement risk, and due diligence action items.

Get same-day alerts when this changes → Get Analyst

Monitor: same-day alerts on the platforms you choose. Analyst: full institutional analysis.

Applicable agencies

  • FTC
    Self-disclosed gaps between public safety representations and actual evaluation practices are relevant to FTC assessment of whether AI safety claims constitute deceptive practices.
    File a complaint →

Provision details

Document information
Document
Anthropic Responsible Scaling Policy
Entity
Anthropic
Document last updated
May 12, 2026
Tracking information
First tracked
July 12, 2026
Last verified
July 12, 2026
Record ID
CA-P-074231
Document ID
CA-D-00823
Evidence Provenance
Source URL
Wayback Machine
Content hash (SHA-256)
9ad6c66902eb2b771ccc40a9d0e1c2d4664c197372e6833be51b902e16754d55
Analysis generated
July 12, 2026 14:38 UTC
Methodology
Evidence
✓ Snapshot stored   ✓ Hash verified
Citation Record
Entity: Anthropic
Document: Anthropic Responsible Scaling Policy
Record ID: CA-P-074231
Captured: 2026-07-12 14:38:45 UTC
SHA-256: 9ad6c66902eb2b77…
URL: https://conductatlas.com/platform/anthropic/anthropic-responsible-scaling-policy/provision/CA-P-074231/self-reported-rsp-compliance-gaps/
Accessed: July 23, 2026
Permanent archival reference. Stable identifier suitable for legal filings, compliance documentation, and research citation.
Classification
Severity
Medium
Categories

Other risks in this policy

Compliance Governance Intelligence

Need to monitor specific governance provisions?

Compliance includes provision-level monitoring, governance timelines, regulatory mapping, and audit-ready analysis.

Arbitration clauses AI governance Data rights Indemnification Retention policies
Get Compliance

Or start with Monitor →

Built from archived source documents, structured governance mappings, and historical version tracking.

Frequently Asked Questions

What does Anthropic's Self-Reported RSP Compliance Gaps clause do?

This provision discloses that Anthropic's adherence to its own stated RSP requirements has not been complete, with specific procedural gaps identified in evaluation timelines, methodology, and buffer assessment; the company characterizes these gaps as minimal risk and uses them as a basis for policy revisions.

How does this clause affect you?

The document discloses that evaluations conducted under the prior RSP version did not fully meet all stated requirements in at least four specific procedural areas, and that the updated policy incorporates revisions intended to address those gaps by adding flexibility and improving compliance tracking.

Is ConductAtlas affiliated with Anthropic?

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