R0042/2026-03-28/Q003/SRC04
OpenAI — Sycophancy in GPT-4o Incident
Source
| Field |
Value |
| Title |
Sycophancy in GPT-4o: What happened and what we're doing about it |
| Publisher |
OpenAI |
| Author(s) |
OpenAI |
| Date |
2025 |
| URL |
https://openai.com/index/sycophancy-in-gpt-4o/ |
| Type |
Vendor incident report |
Summary
| Dimension |
Rating |
| Reliability |
Medium-High |
| Relevance |
Medium |
| Bias: Missing data |
Some concerns |
| Bias: Measurement |
N/A |
| Bias: Selective reporting |
Some concerns |
| Bias: Randomization |
N/A — not an RCT |
| Bias: Protocol deviation |
N/A — not an RCT |
| Bias: COI/Funding |
Some concerns |
Rationale
| Dimension |
Rationale |
| Reliability |
OpenAI is the model provider directly reporting on a known incident. Primary source. |
| Relevance |
Demonstrates that sycophancy is a recognized model provider problem requiring corrective action. However, content could not be fully accessed (403 error). |
| Bias flags |
Some COI — OpenAI has incentive to downplay the severity while demonstrating responsiveness. |
| Evidence ID |
Summary |
| SRC04-E01 |
GPT-4o sycophancy incident demonstrates model provider accountability for anti-sycophancy |