Map one model surface
We pick one production model surface — an API endpoint, a fine-tuned variant, a hosted agent. Local controls hook in next to your existing safety stack and eval gates.
For AI Labs & Model Providers
Runtime controls and signed evidence receipts for the labs powering enterprise AI — so customer security teams, auditors, and regulators see what your model surface actually does, without your data leaving your stack.
Or explore the OVERT 1.0 standard →
Why model cards stopped working
Enterprise customers, auditors, and regulators have stopped accepting model cards and SOC 2 letters as the final word on a frontier model. They want to know what controls run at inference time on the surface they’re actually calling — safety filters, eval gates, policy enforcement, refusal logic — and whether those controls fired on their traffic.
Red-teaming, adversarial testing, and safety evaluations produce real artifacts inside the lab. The gap is at the handoff: those artifacts weren’t designed to be verified independently by a regulated customer, an ISO 42001 auditor, or an EU AI Act notified body.
The Sprint closes that gap on one production model surface in three weeks — runtime controls, signed receipts, and an evidence pack a customer security team can actually verify.
How the sprint runs
Glacis runs inside your infrastructure, instruments one production model surface, and produces signed evidence receipts — without prompts, outputs, or training data leaving your environment.
We pick one production model surface — an API endpoint, a fine-tuned variant, a hosted agent. Local controls hook in next to your existing safety stack and eval gates.
Prompts and outputs stay with you. Runtime controls hash them locally and sign a receipt for each control outcome — only the signed commitment leaves your environment.
An independent notary timestamps the receipts and anchors them in a transparency log. The Sprint closes with an evidence pack your enterprise customer’s security team can verify.
Receipts at the surface
from glacis import RuntimeAssurance a = RuntimeAssurance(workspace="safety-eval-prod") @a.guard( policy="red-team-eval@v3", on_block="escalate") def run_probe(prompt: str) -> str: return model.complete(prompt) # Receipt: local content hash, exported commit, # ECDSA-P256 signature, witness status.
What an evidence pack lets a customer or regulator see
01 / SAFETY
Signed receipts that adversarial prompts and red-team probes were actually evaluated by the production model, at a specific time, with a specific outcome. Auditors verify without ever seeing the underlying prompts.
02 / CARDS
Safety claims link to runtime evidence receipts. “Refusal rate held at X on this benchmark” becomes a verifiable fact, not a marketing line in a PDF.
03 / QUESTIONNAIRE
SOC 2, ISO 42001, EU AI Act, customer security questionnaires — the same evidence pack maps to each control language and ships as the back-up exhibit.
04 / IP
Training data, prompts, outputs, and model weights never leave your environment. Only signed hashes and verification metadata cross the boundary.
Architecture
Not “we don’t store it.” Not “we delete it after.” Prompts, outputs, training data, and weights stay inside your infrastructure — only signed hashes and verification metadata leave.
Related guides
Three weeks. One production model surface. Runtime controls instrumented inside your stack, and an evidence pack a customer security team can verify on their own.
pip install glacis or read the docs →