OpenAI reveals more instances of concerning AI model behaviors during testing
News OpenAI reveals more instances of concerning AI model behaviors during testing Some of its models fabricated information, while others deliberately concealed their unusual behaviors from testers. When it still failed to find the figures, it fabricated them and presented them as facts from a legitimate source. If this had occurred in any other profession, we doubt the perpetrator would have much of a career for long.
- ▪News OpenAI reveals more instances of concerning AI model behaviors during testing Some of its models fabricated information, while others deliberately concealed their unusual behaviors from testers.
- ▪When it still failed to find the figures, it fabricated them and presented them as facts from a legitimate source.
- ▪If this had occurred in any other profession, we doubt the perpetrator would have much of a career for long.
2 outlets in our directory ran this story. All of the coverage we found sits in one bucket: centre. That one-sidedness is itself worth noticing.
Engadget files mainly under tech. We currently carry 722 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Engadget |
| Canonical URL | https://www.engadget.com/2260974/openai-details-instances-of-models-fabricating-information-hiding-from-testers/ |
| Publication time | Thu, 17 Sep 2026 10:50:20 +0000 |
| Retrieval time | 2026-09-17T10:53:44.253Z |
| Last seen | 2026-09-17T10:53:44.253Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | qp2BdZCoHZHW · 2 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
News OpenAI reveals more instances of concerning AI model behaviors during testing Some of its models fabricated information, while others deliberately concealed their unusual behaviors from testers. By Mariella Moon Sept. 17, 2026 6:50 am EST Sean Rayford/Getty Images Add Engadget on Google: Preferred Source Google Discover OpenAI has revealed six incidents, wherein the models it was testing acted on their own and behaved in concerning ways it didn't expect, in a post about how it was adopting a new framework for "misalignment reports." In one one incident, the company said that a model found and used an exposed API key without permission while answering routine questions about earnings figures in a California county.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Engadget.
Discussion
0 commentsMore from Engadget
