The engine
The evaluation engine.
Harm from conversational AI is the same wherever it surfaces, so the instrument has to work everywhere too. The ioLite engine applies one clinical taxonomy at every layer where a person meets a model.
Architecture
One rubric, every layer.
Like an antibiotic, the engine does not care which organ the infection is in. The same taxonomy scores every layer.
13 of 16 components have working code. 4 run end to end in production.
Human–AI interaction layer
Every surface where a person meets a model
Mobile
PartialIn-app SDK and companion app
ioLite Daily ships as a mobile-first app shell with a calm reading experience.
Next · Native iOS and Android apps; an embeddable SDK for third-party apps.
Browser
NextScoring AI chats in the browser
Not offered today.
Next · In-browser scoring for institutions that ask for it.
Direct access
PartialAPI and agent middleware
The audit engine drives target model APIs directly.
Next · Agent middleware that scores tool-using agents in-line.
Enterprise
LiveSchools, clinics, workplaces
Organizations and role-based access in the portal.
Data gathering layer
Capture signals without capturing people
Consent and privacy
LiveOpt-in, on-device redaction
Audits talk to the AI through fictional people, so no real person's conversation is collected.
Signal extraction
LivePsychological safety markers
Every AI reply is scored against the taxonomy, with the words behind each score.
Context tagging
PartialDomain, role, vulnerability
Every finding carries the conversation around it.
Next · Explicit vulnerability and domain tagging.
Model telemetry
PartialWhich model, version, prompt
Audit targets record provider and model for every run.
Next · Model and version capture from live conversations in the field.
Taxonomy and evaluation engine
The core: one rubric, every layer
Taxonomy
LiveSynthesized from validated frameworks
A three-level hierarchy of domains, constructs and sub-constructs, with safety-gate constructs, adopted verbatim from validated clinical frameworks.
Scoring
PartialHarm classes, severity, confidence
Every scenario transcript is scored from 0 to 4 against its construct, with a written rationale that travels with the score.
Next · Calibrated confidence scores and a published clinician agreement study.
Red-team probes
PartialA private, versioned scenario library
A reviewed library of multi-turn scenarios, built per construct and re-run on every model version.
Next · Reviewed scenarios covering every sub-construct.
Certification
PartialAudit reports and safety seal
Graded audit reports with notable failures, plus a grounded assistant that answers questions about each report.
Next · The ioLite certification seal and its renewal cadence.
Compute and infrastructure layer
Where the models actually run
Model endpoints
PartialCloud, on-prem, edge
An encrypted target registry that evaluates cloud-hosted models.
Next · On-prem and edge deployment targets.
Inference monitoring
NextLive scoring at the gateway
Not built.
Next · A gateway service that scores live traffic with the same taxonomy.
Deployment gates
NextBlock, warn, or pass releases
Not built.
Next · Release gates that block a model version that fails safety-gate constructs.
Vendor scorecards
PartialModel-to-model comparison
Audit rollups by construct and strategy, and evaluation seasons for re-runs.
Next · Public model-to-model comparisons, starting with Study 01.
The loop
1. Detect
LiveDiagnose the harm wherever it surfaces
Audits detect failures before release.
2. Neutralize
NextBlock, warn, or reroute in real time
Not built.
3. Immunize
PartialCertify, retest, and keep it from recurring
Re-audits run by season against the same probe sets.
Taxonomy provenance
Built from frameworks, not from opinion.
Assembled from validated clinical and AI-evaluation frameworks, with a source on every node. Sources and definitions are shared under NDA.
3
Priority domains
21
Constructs
69
Sub-constructs
9
Safety gates
Adopted verbatim
Each definition traces back to its source.
Organized into three levels
Domains, constructs, measurable sub-constructs.
Safety gates designated
Where one failure matters regardless of the rest, such as crisis and suicide-risk management.
Mapped to the law
Each construct carries the requirements it bears on, so an audit reads as compliance evidence.
Provenance, by the numbers
4
Source frameworks
93
Taxonomy nodes
9
Safety gates
Domains: Safety, Privacy, and Fairness · Trustworthiness and Usefulness · Design and Operational Effectiveness
Which frameworks, and how the nodes divide between them, is shared under NDA.
The evaluation pipeline
From scenario to evidence.
Every link in the chain leaves a record.
Scenario
Reviewed, multi-turn, written against one sub-construct, re-run on every model version.
Run
The system is taken through it turn by turn. The full transcript is kept.
Score
0 to 4 against the target construct; 3 or above passes. Every score carries a rationale.
Check and report
A clinician can dispute any judgment; we will publish the agreement rate once measured.
Resistance
Why the instrument doesn't decay.
Models change, so won't they learn the test? They will try. The engine is re-run on every model version, so tuning that learns the test instead of the behavior shows up as drift, and each re-run adds labeled transcripts that sharpen the next probes.
Re-audit is structural, not a contract term.
Version-over-version scores
Version-over-version results for the first model family will appear here once Study 01 is published.
Where the engine runs today
Proof at the surface, not a second business.
ioLite Daily shows a calm, carefully measured information product on a phone. The business is the audit and certification behind it.
The list
Get the measurements the day they publish.
One email per release, with a button that opens the full report. Unsubscribe with one click.
Put your system through the engine.
Vendors can request an audit. Institutions deploying AI can talk to us about what the engine measures for them.