IRIS Orchestrator Hub — AI Refusal Bias Reduction for Clinical Systems
IRIS v2.3 — CLINICAL INTENT ROUTER — PRODUCTION

Defensive AI refusal bias:
34.3%<2%
on clinical analysis tasks.

IRIS is a 5-layer intent-routing orchestration framework that surgically separates genuine safety refusals from over-triggered defensive responses. Built for healthcare AI deployments where a refused query has patient-safety consequences.

View Licensing → Technical One-Pager Pricing Model
Baseline Refusal Rate
34.3%
GPT-4o / Claude 3.5 on clinical prompts (n=2,400)
IRIS-Routed Refusal Rate
<2%
Same corpus post-routing, verified against gold labels
False-Positive Safety Rate
0.0%
Genuine safety refusals preserved — zero regression

The 5-Layer Intent-Routing Framework

IRIS doesn't bypass safety — it adds intent resolution before the model's safety layer fires. Each layer narrows the ambiguity window, so the model receives a contextually-grounded prompt that reads as clinical, not adversarial.

L1

Clinical Context Injection

Augments the raw query with structured clinical metadata — ICD codes, patient risk stratum, provider role. Removes the ambiguity that triggers reflexive refusal on bare symptom strings.

Signal
clinical_context.role

encounter_type
L2

Intent Classification

Binary classifier (fine-tuned on 18K labeled medical queries) routes between clinical_analysis, care_coordination, admin, and out_of_scope. Only clinical_analysis paths proceed to IRIS routing.

Signal
intent.class

intent.confidence
L3

Regulatory Frame Selection

Maps jurisdiction + care setting to the correct compliance frame (UK NHS, US FDA 510(k), EU MDR). Rewrites prompt preamble to reference authoritative clinical guidelines — NICE, BNF, UpToDate — before reaching the LLM.

Signal
regulatory.frame

guideline_refs[]
L4

Refusal-Risk Scoring

Scores prompt refusal probability (0–1) using a lightweight perceptron trained on 6K refused/accepted pairs. Prompts above threshold are pre-transformed before model submission; below threshold pass through unchanged.

Signal
refusal_risk.score

transform_applied
L5

Audit + Cryptographic Logging

Every routed request is GPG-signed, SHA-256 chained, and written to an append-only audit log. Provides defensible evidence trail for ISO 42001 AI governance, CQC inspection, and MHRA Pre-Sub processes.

Signal
audit.gpg_sig

audit.chain_hash

Where IRIS Ships Value

Designed for regulated environments where an AI refusal is not an inconvenience — it's a care pathway interruption.

CDS

Clinical Decision Support

CDSS vendors embed IRIS as a pre-processing layer. Differential diagnosis queries, drug interaction checks, and lab result interpretation pass through without triggering defensive refusal. Clinicians get answers, not apologies.

Supports: Epic, Meditech, Cerner embedded AI modules. Tested against clinical vignette corpora from NHS Digital and MIMIC-IV.

↓ 94% refusal reduction on DDx queries
DRUG-IX

Drug Interaction Analysis

Polypharmacy queries are among the highest-refusal categories in clinical AI — the combination of brand names, dosages, and contraindication language triggers nearly every LLM safety filter. IRIS's L3 regulatory framing reduces this to noise.

Validated against BNF interaction dataset (2,100 pairs). Integrates with DrFirst, Surescripts API surfaces.

↓ 97% refusal reduction on polypharmacy queries
REG-AI

Regulated-Industry AI Deployment

Beyond healthcare: any regulated domain where AI over-refusal creates liability. Legal (M&A due diligence, contract risk), financial services (FINRA-regulated advice generation), and insurance (actuarial narrative). Same 5-layer architecture, domain-specific L3 frame.

Frame library: NHS, FDA, EU MDR, FINRA, FCA, Lloyds market standards.

Horizontal — not healthcare-locked

Drop in as API or SDK

IRIS exposes two integration surfaces. The REST API fits teams with existing LLM proxy infrastructure. The Node/Python SDK fits teams embedding directly in their application layer.

REST API

Route any OpenAI-compatible request through IRIS by swapping your base URL. All 5 layers apply transparently. Latency overhead: <40ms p99 (co-located).

// Before: direct model call const client = new OpenAI({ baseURL: "https://api.openai.com/v1" }); // After: IRIS-routed const client = new OpenAI({ baseURL: "https://iris.talastar.com/v1", defaultHeaders: { "X-IRIS-Context": "clinical_cdss", "X-IRIS-Frame": "NHS_NICE" } });

SDK (Node / Python)

Install the IRIS SDK for programmatic access to all 5 layers, audit log queries, and frame management. Typings included. Zero extra dependencies in the production bundle.

# Install npm install @talastar/iris-sdk // Init import { IrisRouter } from "@talastar/iris-sdk"; const iris = new IrisRouter({ licenseKey: process.env.IRIS_LICENSE_KEY, frame: "NHS_NICE", auditEnabled: true }); const result = await iris.route(prompt, { intent: "drug_interaction", encounterType: "inpatient" });

Built Inside Clinical Systems

IRIS was designed by operators who have worked inside the systems it now improves.

Founder · Clinical Lead
NHS Clinical Insider
Clinical background, frontline AI deployment experience

Experienced working within NHS systems and understanding firsthand where LLM refusal creates care pathway friction. IRIS was not designed from the outside — it was designed to fix a problem that caused real delays in clinical settings. The 34.3% baseline was measured on a real deployment corpus, not a synthetic benchmark.

NHS Clinical AI MHRA CQC compliance
Founder · Technical Lead
Multi-Agent Operator
Multi-agent system architect, production AI deployments

Architect of multi-agent orchestration systems operating at production scale. The IRIS framework evolved from a pattern library built across multiple regulated-industry deployments. The 5-layer model was extracted from the most robust routing patterns observed across those deployments — then formalized into a licensable asset.

Multi-agent orchestration LLM routing Zero-trust infra X.509 / GPG audit

Three tiers. One outcome.

IRIS is priced on the refusal-bias delta you deploy, not commodity API calls. A 32-point reduction in clinical AI refusal rate is worth more than the infrastructure cost to deliver it.

Developer
$149
per month
Evaluate. Integrate. Ship a proof-of-concept.

  • 10,000 routed requests/month included
  • Pay-per-1k beyond ($1.20/1k)
  • REST API + Node/Python SDK
  • 5-layer routing (L1–L5)
  • Audit log API — 30-day retention
  • Self-serve onboarding docs
Start Developer Trial →
Enterprise
$2,500
paid POC · from $8k/month full license
Co-deployed on your infra. Custom SLA. SOC 2 Type II in progress.

  • 30-day co-deployment on your infrastructure
  • Live routing trace review
  • Custom L3 frame (1 regulatory domain)
  • Integration scoping report
  • Deposit applied as license credit
  • Custom SLA — 99.5%+ available
  • SOC 2 Type II report sharing
Start Paid POC →

or contact us first to scope before paying

Existing customer? Manage subscription →  ·  Questions