Skip to main content

Intent Model (Preview)

In development. This entry describes direction and engineering progress, not a shipped, generally available feature. No detection-accuracy claims are made here.
Quint’s behavioral approach reads what an AI agent says it is doing and checks whether that intent licenses what it actually does. The Intent Model is the on-device component that scores that relationship: given a stated intent and a concrete action, does the intent license the action, or does the action exceed it.

What this is building toward

  • On-device, privacy-first. The model runs at the endpoint. Conversation content stays on the endpoint. Only structured, de-identified geometry is ever used for learning, never raw text.
  • The model proposes, the scope disposes. The Intent Model never makes an enforcement decision on its own. Its judgment shapes a scope that is then enforced deterministically, alongside fixed sensitivity floors, policy ceilings, and a kernel truth channel the agent cannot author. Fooling the model still leaves an attacker inside those deterministic limits.
  • Calibrated per policy key. Operating thresholds are calibrated per action class and sensitivity, not by a single global setting, so the decision boundary reflects the real risk of each capability.

Engineering progress

  • Established a leak-free evaluation methodology for intent-to-action licensing. The instrument is constructed so that a model cannot appear accurate by reading the action shape alone; it must judge the actual relationship between intent and action. Rigorous, gameable-metric-resistant evaluation is a foundation we are building before making capability claims.
  • Confirmed an on-device, no-GPU training path for the everyday model update, keeping the learning loop cheap and the endpoint footprint bounded.
  • Upgraded the on-device encoder to a larger, more capable model, chosen and evaluated on the same leak-free instrument. It continues to run at the endpoint, per turn and asynchronously (off the blocking path), and the everyday model update remains a cheap, GPU-free retrain. No conversation text leaves the endpoint.

Status

Pre-general-availability. We will describe measured capabilities here once they are validated on the leak-free evaluation at scale. See the platform roadmap (Adaptive) for how the Intent Model fits the broader release plan.