Independent public-interest evaluations of how systems respect dignity and support agency.
HXR Evaluation
ChatGPT Read Aloud Playback Continuity and Audio Integrity
HXR is evaluating whether ChatGPT Read Aloud maintains stable, intelligible audio throughout continuous playback. One contributor reports progressively worsening static after roughly 60–90 seconds of continuous listening, with clean playback temporarily restored by restarting or reseeking.
- Evaluation ID
- UXR-EVAL-0014
- Status
- Open — gathering evidence
- HXR Incident connections
- 1
- Last updated
- August 18, 2026
Who and what this Evaluation covers
Naming an organization, product, service, or workflow identifies the subject of this Evaluation. It does not by itself prove which internal team caused or controls the issue.
- Organization in scope
- OpenAI
- Product, service, or workflow
- ChatGPT — Read Aloud
Evaluation scope
User-facing ChatGPT Read Aloud playback continuity and audio integrity in the contributor-observed mobile-app context. The Evaluation includes progressive static onset, elapsed-time-from-start/resume behavior, restart/reset recovery, and listening burden. It does not claim a specific internal technical cause, product-wide prevalence, or behavior across all ChatGPT platforms.
Question being evaluated
Does ChatGPT Read Aloud maintain stable, intelligible audio quality throughout continuous playback without progressively accumulating static or forcing the listener to restart playback?
Current findings
Difficulties and opportunities to improve
Current evidence supports a bounded contributor-reported finding that ChatGPT Read Aloud developed progressively worsening static after roughly 60–90 seconds of continuous playback and required restarting playback to temporarily restore clean audio.
Mixed findings and limits
Restarting or reseeking temporarily restores clean playback for the contributor. The onset interval is approximate, and exact ChatGPT app version, iOS version, audio-output route, prevalence, and internal mechanism remain unresolved.
Organization information and response status
No response requested from or received from OpenAI.
How HXR would test performance
Play multiple sufficiently long ChatGPT Read Aloud responses continuously and begin additional runs from different seek positions. Audio should remain intelligible and free of progressively increasing static for the full playback duration, and restarting or reseeking should not be required to preserve audio quality.
Evidence coverage and limits
Current support consists of one bounded recurring-condition Incident, UXR-2026-0818-0001, and its preserved contributor testimony. The contributor reports a restart/seek observation that separates session duration from fixed content position, plus clean same-device use of Gemini, Claude, Copilot, podcasts, Apple Music, and SiriusXM. These controls narrow the observed user-facing pattern but do not independently establish causation or prevalence.
Connected HXR Incident reports
- Supports: UXR-2026-0818-0001 — ChatGPT Read Aloud Develops Progressive Static During Continuous Playback
1 public HXR Incident report is connected to this Evaluation. Canonical roles: 1 supports; 0 partially supports; 0 narrows; 0 context only; 0 challenges. Connections are not votes and do not establish prevalence.
Next observation or verification
Add independent reproductions, version and audio-route observations, contradictory evidence, and any verified implementation change or reform result while preserving the original Incident and testimony unchanged.
Revision history — Public Revision 1
Public Revision 1 creates the living Evaluation and links its first supporting public Incident, UXR-2026-0818-0001.
Technical record details
- Evaluation type
- Product or Application
- Benchmark posture
- Developing Evidence
- Blue Score readiness
- Not Ready
- Reform / positive practice
- No Current Reform Work
- Public revision
- 1