Is your organisation ready for the autonomy it intends to give AI?

AI is moving from assisting work to initiating, coordinating and executing it. That changes authority, accountability and trust at work. Sociotek shows whether your organisation can support the autonomy you plan, and aligns the people who must make it work.

Request early access to the PulseSee how the diagnostic works

The technology is arriving faster than the organisation can receive it

Corporate AI initiatives begin with models, licences, pilots and use cases. The organisations expected to absorb them still run on human relationships, negotiated authority, tacit knowledge, incentives, trust and accountability.

When agents enter that system they redistribute who, or what, may interpret information, initiate action, make decisions and use the organisation's resources. That is a question of organisational design before it is a question of technology.

Five gaps recur. Technology can do work the organisation cannot yet delegate, supervise or recover. Leaders believe strategy and communication are clearer than employees experience them. Business and technical teams lack a shared view of the work. A successful pilot is mistaken for a scalable way of working. And maturity surveys produce averages, where what is needed is agreement across functions on what will change.

61%of leaders expect most of the AI agents they use to be generally autonomous within four years
5%say their business processes are highly prepared for AI agents
25%say their workforce is prepared, the lowest of any area except business processes

Source: Deloitte, AI readiness gap survey, 12 August 2026.

One system, nine dimensions

The Sociotek diagnostic treats readiness as a property of the whole system of work. Six dimensions describe the organisation that must absorb the change. Three describe the operating conditions that agents themselves require. Weakness in one cannot be averaged away by strength in another.

Warm marks the human system: judgement, relationships, accountabilityCool marks the agent system: identity, permissions, execution

The enduring organisational core

  1. Strategic leadership alignmentShared intent, agreed limits on delegation
  2. Workflow and value readinessThe work understood end to end
  3. Work, role and decision designWho decides, who owns, what changes
  4. Trust, participation and employee experiencePeople informed, involved and treated fairly
  5. Learning and adaptive capabilityJudgement protected and developed
  6. Change and scaling capacityManagers equipped, criteria to scale or stop

The agentic operating core

  1. Agent infrastructure, identity and orchestrationEvery agent known and scoped
  2. Governance and responsible autonomyAutonomy matched to consequence
  3. Operational assurance and resilienceTrace it, stop it, reverse it

The nine dimensions explained in plain language

Readiness is measured against the autonomy you intend

An organisation ready for an AI adviser may be far from ready for an agent that initiates its own work. The diagnostic asks what level of autonomy you plan for a specific workflow, then tests readiness against that level. Few workflows belong at the top.

A0Human executionAI retrieves or generates information. People act.
A1AdviserAI recommends. A person interprets, decides and acts.
A2Approved executorAI acts within bounds after explicit human approval.
A3Bounded operatorAI completes a defined workflow and escalates exceptions.
A4Persistent delegateAI pursues continuing goals and initiates work.
A5Orchestrated systemMultiple agents divide work and hand off tasks.

Up to A2 a person approves each action. From A3 the agent acts first, and a set of controls becomes a condition of proceeding. The six levels, with everyday examples

Questions people ask first

What is human-agent operating readiness?

It is an organisation's ability to allocate purpose, work, authority and accountability between people and AI agents responsibly. It covers leadership alignment, workflow and role design, employee trust, learning, change capacity, agent identity and infrastructure, governance and operational assurance.

How is Sociotek different from an AI maturity assessment?

Most AI maturity assessments score technology, data and governance and report an average. Sociotek tests readiness against the specific autonomy you intend, treats critical controls as conditions that strengths elsewhere cannot offset, compares the views of different stakeholder groups, and follows the assessment with a facilitated alignment session.

What is the Pulse?

The Pulse is a short version of the diagnostic for one person to complete in under ten minutes. It returns a nine-dimension profile and three questions to take to your leadership team. It is in preparation, and you can register for early access.

Is the diagnostic validated?

The instrument is research-informed and under validation. Sociotek describes it as a structured readiness conversation until its testing programme supports stronger claims.

Start with a conversation

Tell us what you are deciding. Ask for early access to the Pulse, a founding partnership, a consulting conversation or a keynote, and Dennis will reply personally.