Match authority to consequence
Decision rights increase with capital at risk, irreversibility, stakeholder impact and operating consequence.
DNA TECH GROUPContact Governance
Clear authority.
Visible evidence.
We are committed to driving AI governance and safety through a single, integrated cross-functional standard—with clear authority, shared evidence and human accountability across every business.
01
Decision principles
Keep assumptions visible
Material forecasts, dependencies, exclusions and approval conditions remain connected to the decision they support.
Challenge outside the origin team
Technical, risk, security, ethics and resilience questions receive review with enough independence to matter.
02
Decision rights
Evidence required at the moment of commitment.
| Decision | Required basis | Accountability |
|---|---|---|
| Opportunity entry | Strategic fit, demand signal, fatal-flaw screen and named sponsor | Investment leadership |
| Development advance | Site, resource, stakeholder, technical and commercial evidence | Stage-gate committee |
| Capital commitment | Mature risk position, delivery baseline and accountable operating owner | Authorized decision body |
| Operating change | Safety, reliability, security, environmental and service implications | Platform operations |
03
Control environment
Controls organized around the way risk enters.
Technical assurance
Design basis, independent review, change control, commissioning evidence and operating readiness.
Responsible technology
Use-case classification, data rights, evaluation, human authority, security and ongoing monitoring.
Environmental + community
Resource limits, impact pathways, engagement commitments, grievance routes and performance evidence.
Commercial integrity
Conflicts, partner diligence, approval authority, records, representations and transaction controls.
04
AI governance + safety
One integrated standard. Shared responsibility.
Our commitment is to unite engineering, cybersecurity, privacy, legal, risk, compliance and operations around a common standard for responsible AI. We aim to make safety part of every decision, from the first use-case assessment through retirement.
Define purpose and accountability
Establish the intended use, affected people, risk level and accountable owner. Bring the relevant functions together to agree on data rights, acceptable use and approval conditions.
Evaluate before deployment
Assess reliability, bias, privacy, security and foreseeable misuse in the operating context. Record limitations and require review proportionate to the consequences of failure.
Preserve human authority
Define who can approve, challenge, override or stop an AI system. Set escalation paths and safe fallback procedures for consequential decisions.
Monitor and improve
Track performance, incidents and changes throughout the lifecycle. Use shared evidence to reassess risk, strengthen controls and decide when to restrict or retire a system.
This is our commitment to a common operating standard across functions and business domains, with controls adapted to each use case and its local requirements.
Discuss AI governance and safetyQuestions + concerns
Raise the issue where it can be acted on.
Governance, ethical, security or integrity concerns can be directed to the corporate team for appropriate review and routing.
governance@dnatechgroup.com