TypeSafe AI: Fast, Typed AI Decisions for Security Automation

TypeSafe AI introduces Jev, a "System One" model designed for fast, narrow security judgments rather than text generation. Unlike LLMs that return free-form text requiring parsing, Jev accepts state data plus typed questions and returns typed answers—probabilities, choices, or scores—with calibrated confidence via RLCD training. The article argues that using large language models for quick AppSec decisions (like reviewing agent skills or triaging bug reports) is slow, expensive, and prone to output-parsing failures and prompt injection. Jev runs in roughly 100 ms at $42 per billion input tokens, operates without tools or text output, and is positioned as a fast judge inside code that escalates uncertain cases to humans or reasoning LLMs. The post walks through three AppSec use cases—catching malicious agent skills, bug bounty triage, and agent tool-call guardrails—and details installation, SDK usage, context-window limits, known weaknesses (adversarial input can still shift answers), and adoption guidance such as keeping deterministic rules and thresholds in code, pinning model versions, and red-teaming questions before production use. 

https://benhurott.com/ai-security/appsec/2026/09/19/typesafe-ai-system-one-for-appsec.html

Comments

Post a Comment

Popular posts from this blog

Prompt Engineering Demands Rigorous Evaluation

In the blog post “Prompt Engineering Requires Evaluation” on the Shostack + Associates website, the author argues that treating prompts for large language models (LLMs) merely as creative artefacts is insufficient. Engineering prompts properly demands structured evaluation frameworks — what the AI community calls “evals” — to test which prompt versions work better, with which models, and under which conditions. The post highlights that simply assuming a prompt is “good enough” creates risks when LLMs are integrated into production systems (e.g., for threat modeling). It advocates for measuring prompt performance, variation effects, and tool-chain dependencies (model, context, ancillary materials).  Ultimately the message is: prompt engineering should borrow disciplined practices from software engineering (versioning, testing, benchmarking) rather than relying on informal experimentation.  https://shostack.org/blog/prompt-enignieering-requires-evaluation/
Read more

OWASP ASVS 5.0 Released - Key Updates and What You Need to Know

The OWASP Foundation has released version 5.0 of the Application Security Verification Standard (ASVS), a major update to their security framework for web applications. This new version features restructured security requirements for better clarity, expanded guidance for cloud and API security, improved DevSecOps integration for CI/CD pipelines, updated threat modeling support, and enhanced compliance mapping with standards like NIST and PCI DSS. ASVS serves as a critical benchmark for developers building secure applications, penetration testers conducting security assessments, and auditors performing compliance reviews.  The standard is available for download from the OWASP ASVS project page, with organizations encouraged to integrate it into their software development lifecycles through code reviews and security testing tools. As a vendor-neutral, community-driven project, OWASP continues to welcome contributions to further develop the standard. This release represents an importa...
Read more

Open-SPDD proposes an open framework for Spec-Driven Development workflows

Open-SPDD is an open-source project focused on Spec-Driven Product Development, aiming to formalize how software specifications, requirements, architecture, and implementation interact throughout the development lifecycle. The project promotes structured specification management, traceability, and AI-assisted engineering workflows, positioning itself as a foundation for “SpecOps”-style development practices. Its GitHub materials describe SPDD as a way to reduce ambiguity between planning and execution by treating specifications as first-class operational artifacts that can drive automation, validation, and collaboration across engineering teams  https://github.com/gszhangwei/open-spdd
Read more