Operational Clarity produces independent AI Authority Assessments for organisations seeking an evidence-based understanding of what AI agents and AI-enabled platforms are documented as able to do.
Research is not limited to platforms already covered by the SoftTech AI Registry. Where a platform is already covered, that published research is a starting point. Where it is not, research is carried out from documentary sources.
Research into the documented AI capabilities and authority of one platform or agent product.
Research across several AI-enabled platforms, including where their documented authority overlaps or combines.
Introducing AI agents or AI features into existing platforms
Reviewing what AI tools are permitted to do on the organisation's behalf
Governance, risk or board discussions about AI use
Procurement or renewal of AI-enabled software
What vendor documentation states an agent or AI feature can do.
Which actions are permitted, and whose authority the AI acts under — including documented user permissions, service accounts, application identities or other relevant mechanisms — and how that authority is documented as being granted or limited.
Where documented capabilities extend to external systems, people or services.
Where documentation states a person must confirm an action, and where it does not. Where the available evidence supports it, the assessment distinguishes platform-enforced approval, configurable approval and procedural approval.
Each engagement produces a structured analytical report setting out documented AI capabilities, action authority, permission boundaries, human confirmation requirements, supporting evidence and unresolved questions.
For multi-platform engagements, the report may examine documented authority relationships across products. Vendor documentation alone does not establish how an organisation has configured or operates those products.
Every commissioned engagement includes:
A defined, written scope
A stated research cut-off date
Documented sources
Stated evidence limitations
Agreed deliverables
Commissioned work does not provide certification, endorsement, privileged vendor access, penetration testing or continuous monitoring. Research into documented vendor capabilities is separate from examining an organisation's own implementation. Any review of client-supplied configuration evidence must be explicitly agreed and its verification limitations stated.
Paying for research cannot buy favourable findings or changes to published research conclusions. Commissioned reports are treated as confidential under the agreed engagement terms. Published research remains independent of commercial engagements.
Engagements are individually scoped and quoted. Fees are agreed before work begins.
Enquiries are received through a single private form on the Services page. Scope, feasibility and evidence availability are discussed before any work is agreed.