Jev Is Not a Language Model, but It Breaks Like One
Share A new kind of AI model returns decisions instead of text. In this article Jev is a new kind of AI model that returns typed decisions instead of text, built for software to call rather than for people to read. A model that writes text gets checked by the person reading the text.
- ▪Share A new kind of AI model returns decisions instead of text.
- ▪In this article Jev is a new kind of AI model that returns typed decisions instead of text, built for software to call rather than for people to read.
- ▪A model that writes text gets checked by the person reading the text.
2 outlets in our directory ran this story, first to last over 9 hours. All of the coverage we found sits in one bucket: centre. That one-sidedness is itself worth noticing.
- ▪ Contrastive Language Models — Notion
Hacker News (Front Page) files mainly under programming. We currently carry 2,091 of its stories. Top-voted stories on Hacker News.
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 | Check Point Blog |
| Canonical URL | https://blog.checkpoint.com/ai-security/jev-is-not-a-language-model-but-it-breaks-like-one-prompt-injection-against-a-typed-decision-model/ |
| Publication time | Thu, 24 Sep 2026 13:16:46 +0000 |
| Retrieval time | 2026-09-24T15:10:26.286Z |
| Last seen | 2026-09-24T15:10:27.825Z |
| 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 | U83D3hnIfh7M · 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
Share A new kind of AI model returns decisions instead of text. We spent a day trying to change its mind. It cost about 50 cents. In this article Jev is a new kind of AI model that returns typed decisions instead of text, built for software to call rather than for people to read. We put it inside a realistic application and tried to change its verdict from the outside. →every configuration we tested was broken: risk downgraded to low and investment advised, on a document that flags every warning sign →about 50 cents per successful break, with our strongest attacker landing on the fourth turn on average →typed, structured input did not make the model meaningfully harder to manipulate, and neither did telling it to distrust the document →reasoning was the strongest defense we measured, and…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Check Point Blog.
Discussion
0 commentsMore from Check Point Blog
