Article
CommunityJev Media & Talks
InfoWorld: Jev 'even more susceptible to attack' than LLMs
InfoWorld covers a HiddenLayer report finding Jev more vulnerable than traditional LLMs to adversarial text and prompt injection, with analysts calling the methodology sound but debating whether flaw reports this early are premature.
- Format
- Article
- Reading time
- 2 min
- Published by
- Community
- Author
- Evan Schuman
- Last verified
- Oct 2026
Used inJev for Security & Privacy
Highlights
- HiddenLayer's blog post says Jev is in some cases even more vulnerable to attacks, citing adversarial text attacks and prompt injection.
- Security research director Kasimir Schulz: Jev is great, you should use it, and it has a lot faster response time.
- Schulz says TokenBreak normally does not work with larger LLMs, but Jev is vulnerable to it out of the box.
- TypeSafe did not respond to multiple requests for comment, the article reports.
- Gartner's Nader Henein says being surprised Jev has issues is like being surprised when your baby drools.
Watch out
Copy reviewed against the published article on 2026-10-17. The underlying report is HiddenLayer's, a security vendor that sells AI security products, a point analysts in the article call somewhat self-serving, and TypeSafe's response is unrecorded.
Questions about this entry
Short answers drawn from the linked project or page.
That Jev can be even more vulnerable to attacks than traditional LLMs, with adversarial text attacks and prompt injection able to disrupt the model's outputs, resulting in incorrect and potentially devastating actions.
Kasimir Schulz says Jev is great and you should use it because of its faster response time, with the caveat that it does not have all of the protections seen in traditional LLMs, so some attacks succeed that should have failed.
TokenBreak inserts characters in front of important words to evade classifiers. Schulz says it normally does not work with larger LLMs, but Jev is vulnerable out of the box, and HiddenLayer found heavier, better-targeted perturbation raised success rates.
The article says TypeSafe did not respond to multiple requests for comment. It also notes debate over whether Jev is technically an LLM, with TypeSafe suggesting it is not and HiddenLayer saying it appears to be one.
IDC's Philip Harris called the red-teaming methodology sound and consistent with established practice; Gartner's Nader Henein called the reports premature for a month-old model; Dickson Research's Frank Dickson said speed is a feature, not a security control.
Analysts agreed with both the methodology and technical conclusions while pointing out that a security vendor concluding a new approach is not secure is somewhat self-serving, since HiddenLayer sells AI security products.
That constraining the format of a decision does not constrain its correctness or consequences, that the same speed making Jev attractive also makes wrong decisions faster and cheaper, and that no model should be trusted to stay within its guardrails.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
YouTube: Kubernetes admission control with a decision model, fail-closed.
Hacker News: SafeDep Gryph plus Jev checks for compromised coding-agent behavior.
2 pointsDetecting Compromised AI Coding Agents with Jev and GryphRead the thread on Hacker NewsX: Jev and Laya on security queries
X: Testing Jev for prompt injection
More Jev resources like this
The Register covers HiddenLayer's tests of Jev and a decision-tuned Qwen model: a planted ad stole a browser agent's click on a simulated store page, and tokenbreak edits snuck toxic prompts past verification. TypeSafe did not immediately respond.
Jev Media & TalksArticle#community#press#security
Community
The Injection's brief on HiddenLayer's tests: a browser agent powered by jev-ultrafast followed instructions planted in a web page and picked an attacker's ad over the real checkout button, and TokenBreak edits let toxic prompts slip past Jev's scoring. The post records no fix or response from TypeSafe.
Jev Media & TalksArticle#community#press#security
Community
VentureBeat's Louis Columbus reports Jev is spreading into agent infrastructure faster than enterprises can audit it. In one Octomind test a planted field cut a block probability from 0.76 to 0.48, and TypeSafe and Pydantic both warn inputs can steer answers.
Jev Guides & ArticlesArticle#community#press#security
Community
A Spanish-language analysis of Jev's five-day rise to critical infrastructure, the documented injection finding that moved a block verdict, and the six controls it says to add before production.
Jev Media & TalksArticle#community#press#security
Community
Cisco's security scanner for agent skills added a System One analyzer: it sends the full skill text to POST /v1/systemone as one advisory Noul question and records the probability, while being deliberately unable to emit a finding or change a severity.
Jev Tools & CLIs#community#security#agent-skills
Community
cisco-ai-defense/skill-scannerPython
cisco-ai-defense/skill-scannerPythonPython CLI and agent skill that reviews Agent Skills, MCP configs, and source code for suspicious behavior with TypeSafe Jev plus local static checks, reporting file and line evidence, risk categories, model probabilities, and unscanned scope.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
win4r/jev-security-scanPython
win4r/jev-security-scanPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.