Provision record
OpenAI · GPT-4o System Card (PDF) · View original document ↗

Speaker Identification Refusal Mitigation

Medium severity High confidence Explicit document language Unique · 0 of 352 platforms
Stay ahead of the changes
Track OpenAI and get the diff the day its terms change.
Share 𝕏 Share in Share 🔒 PDF
Document Record

What it is

GPT-4o is post-trained to decline requests to identify private individuals or celebrities based solely on their voice in audio input, while retaining the ability to attribute famous historical quotes to known speakers.

This analysis describes what OpenAI'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 a post-training behavioral restriction targeting voice-based biometric identification, addressing privacy and surveillance risks associated with audio input processing in a real-time conversational AI context.

Clause Stability Stable

0
Changes
5
Months Monitored
Jul 9, 2026
First Seen
Jul 9, 2026
Last Seen

Consumer impact (what this means for users)

Under this provision, GPT-4o will decline to identify a person based on their voice in an audio input in the large majority of cases, with the document reporting a 98% correct refusal rate for should-refuse scenarios in the deployed model, up from 83% in the early model.

Cross-platform context

See how other platforms handle Speaker Identification Refusal Mitigation and similar clauses.

Compare across platforms →
▸ View Original Clause Language DOCUMENT RECORD
"
We post-trained GPT-4o to refuse to comply with requests to identify someone based on a voice in an audio input, while still complying with requests to identify people associated with famous quotes. GPT-4o still complies with requests to identify famous quotes. For example, a request to identify a random person saying four score and seven years ago should identify the speaker as Abraham Lincoln, while a request to identify a celebrity saying a random sentence should be refused.

Excerpt from OpenAI's GPT-4o System Card (PDF)

ConductAtlas Analysis

Institutional analysis (regulatory & governance intelligence)

(1) REGULATORY LANDSCAPE: Voice-based identification from audio input engages biometric data protection frameworks including the Illinois Biometric Information Privacy Act (BIPA), GDPR Article 9 on special category biometric data, and the EU AI Act's provisions …

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 →
  • State Attorney General
    State AGs in California, New York, Texas, and other states can investigate violations of state consumer protection and privacy laws, including CCPA (California), SHIELD Act (New York), and equivalents.
    Who can file: Residents of states with comprehensive privacy laws — primarily California, Virginia, Colorado, Connecticut, and Utah
    What you need: Evidence of the violation, explanation of how your state rights were affected, and your account or contact information with the company
    What to expect: Outcomes vary by state. May result in investigation, enforcement action, or requirement for the company to change practices. No direct individual compensation in most cases.

    Search "[your state] attorney general consumer complaint" to find your state's direct complaint form

Provision details

Document information
Document
GPT-4o System Card (PDF)
Entity
OpenAI
Document last updated
March 5, 2026
Tracking information
First tracked
March 10, 2026
Last verified
July 9, 2026
Record ID
CA-P-013520
Document ID
CA-D-00008
Evidence Provenance
Source URL
Wayback Machine
Content hash (SHA-256)
45119e7807db6253470f10984f48c3d17616fd77831d8ea69c793524d7ec99b4
Analysis generated
March 10, 2026 03:40 UTC
Methodology
Evidence
✓ Snapshot stored   ✓ Hash verified
Citation Record
Entity: OpenAI
Document: GPT-4o System Card (PDF)
Record ID: CA-P-013520
Captured: 2026-03-10 03:40:55 UTC
SHA-256: 45119e7807db6253…
URL: https://conductatlas.com/platform/openai/gpt-4o-system-card-pdf/provision/CA-P-013520/speaker-identification-refusal-mitigation/
Accessed: Aug. 18, 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 OpenAI's Speaker Identification Refusal Mitigation clause do?

This provision establishes a post-training behavioral restriction targeting voice-based biometric identification, addressing privacy and surveillance risks associated with audio input processing in a real-time conversational AI context.

How does this clause affect you?

Under this provision, GPT-4o will decline to identify a person based on their voice in an audio input in the large majority of cases, with the document reporting a 98% correct refusal rate for should-refuse scenarios in the deployed model, up from 83% in the early model.

Is ConductAtlas affiliated with OpenAI?

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