For organisations deploying conversational AI where psychological risk may emerge
Runtime safety and governance middleware for AI conversations.
CommonLayer sits between your application and the AI model. It detects risk in real time, enforces your safety policies, escalates to human care when it matters, and records every decision in an auditable log.
Built by clinicians and engineers who design safety into conversational systems from the outset, CommonLayer is currently being validated with clinical, enterprise, and governance partners.
Designing for the Failure That Matters Most
The dangerous failure is not the missed signal.
It is the harmful response.
In mental health the costly error is rarely the risk you miss. It is the conversation that gets cut off or dropped into a generic crisis script the moment distress appears. CommonLayer is built to do neither. It adapts the model's behaviour in real time, keeps the conversation supportive and safe, and escalates only when your policy and the context call for it.
It protects the person rather than abandoning them.
Independent by Design
Foundation models build capability.
CommonLayer builds the trust around it.
Enterprise trust needs independent oversight. You would not let a company audit itself, and high-risk AI is no different. That's why CommonLayer runs across foundation models rather than inside any one of them, applying your policies, validating behaviour, and keeping an independent, auditable record.
Whichever model powers the conversation, governance should not belong to its vendor. Trust, policy, accountability, and compliance are enterprise responsibilities that must remain independent of the underlying AI model.
Built in Partnership. Guided by Transparency
CommonLayer is currently in development, and we believe trust begins with being transparent about where we are today and where we are going. We are working closely with clinicians, researchers, enterprise partners, and governance experts to validate the platform, refine its capabilities, and ensure it meets the real-world demands of deploying conversational AI safely at scale.
As the platform evolves, we will openly communicate what is production-ready, what is being evaluated with design partners, and what remains on our roadmap, alongside our approach to privacy, security, governance, and performance. We see transparency and partnership not simply as product principles, but as the foundation for building the trusted infrastructure that enterprise AI requires.
The Trust Infrastructure for Conversational AI
Detect risk in context
Evaluate configured psychological and behavioural indicators across the permitted conversational window.
Manage Human Escalation
Route qualifying events into approved clinical, EAP, crisis or safety workflows.
Apply runtime controls
Change response policies, tools, boundaries and continuation rules when thresholds are reached.
Create governance evidence
Record the classification, threshold, policy, model version, action and escalation result.
OUR APPROACH
A Partnership, Not Just Another Platform
Every pilot begins with jointly agreed safety, integration and operational success criteria. We evaluate detection performance, policy execution, escalation completion, system latency and governance coverage against those criteria before production deployment.
About our team
Built by people who have done this before.
CommonLayer is clinically led. Founder Chris Rhyss Edwards has spent more than twenty years building digital systems that shape behaviour at scale, across News Corp, Clemenger, and a decade leading an internationally awarded healthtech company, and is completing doctoral research into safer AI for mental health.
He is joined by a Clinical Science director, a CTO building the runtime layer, UX and design thought leaders, and supported by an experienced CMO and and a legal advisor on governance.
Chris Rhyss Edwards
CEO & Founder
Nicki Kenyon
Non-Exec Director & CMO
Lucas Sargent
Chief Technology Officer
Dr Louise Metalf
Director Clinical Science
Oliver Weidlich
Advisor & UX/UI Lead
Stratos Karousos
Legal Advisor
Schedule a call
Discover how CommonLayer helps your organisation deploy conversational AI with confidence. Whether for employee wellbeing, customer engagement, or enterprise productivity, we'll show how CommonLayer provides the governance, behavioural intelligence, and compliance needed to build trusted AI at scale.
Chris Rhyss Edwards
CEO at CommonLayer AI
CommonLayer AI
Build trusted conversational AI with enterprise-grade safety, governance, and compliance.
Resources
Legal
Contact
info@commonlayer.ai