Safe AI foundations
Appropriate use, restricted information, prompting, output review, hallucination risk and the limits of automation.
Output: AI-readiness assessment
AI training + operating knowledge
Train one team to use AI safely—and capture how their work actually happens.
Run a focused four-to-six-week pilot with one department. Employees build practical, role-specific AI skills while creating manager-approved role playbooks, workflow maps, AI guardrails and prioritised automation opportunities.
The result is not just completed training. Your organisation keeps a structured operating asset for onboarding, process improvement, governance and future AI implementation.
- 5–10 participants
- One department
- Manager-approved outputs
The operational gap
AI training alone does not solve the operational problem.
Employees are already experimenting with AI, but many organisations still lack a clear view of the tools, information, tasks and human checks involved—or of how important work moves between people and systems.
TIQ+ connects learning to real work. Employees apply AI to genuine activities, managers review what they produce, and the organisation retains the approved outputs.
Which AI tools employees are using—and for which tasks
What company information is being entered
Where human review is required
Which processes are suitable for automation
How important work moves between people and systems
See one workflow become an approved company asset.
Follow a realistic new-client-onboarding workflow from current practice to a controlled AI opportunity and manager-approved operating asset.
Example stages
Example progress
01 / current workflow
Employee draftNew client onboarding
Client services manager
Signed engagement agreement
Account managers rely on personal experience to remember record creation, checks, document requests, review steps and information restrictions.
Repeated manual drafting
Inconsistent checks and handoffs
02 / role and workflow capture
Structured captureDocument how the work happens
Account manager
CRM · document store · email
- 01Confirm agreement and owner
- 02Create the CRM record
- 03Complete data and conflict checks
- 04Prepare the welcome communication
- 05Review the client pack
- 06Send and schedule kickoff
03 / AI opportunity
Proposed useTarget one repeatable activity
Prepare a first draft of the welcome communicationUsing an approved AI tool and non-confidential information
Hightask frequency
Stableprocess pattern
Reviewhuman oversight
Reduce repetitive drafting time.
Proceed only with defined controls.
04 / required controls
Draft guardrailsDefine the boundaries before use
Information restrictionNo confidential client information in an unapproved AI tool.
Human verificationThe account manager checks all names, facts and disclosures.
Quality benchmarkUse the approved communication template as the reference.
AccountabilityThe account manager remains responsible before sending.
05 / manager review
Manager viewCorrect, control and approve
JD
James DoyleAccount director · authorised reviewer
Ownership confirmedControls approved
Only approved content enters the organisation’s operating-knowledge library.
06 / approved company asset
Manager approvedA reusable onboarding workflow
Client services / New client onboardingVersion 1.0 · authorised company knowledge
Draft a welcome communication with non-confidential information.
Account manager verifies and approves before sending.
Documented workflow · named ownership · AI use case · information restrictions · human-review rules · quality benchmark
Step 1 of 6 · choose any stage
This is a fictional example showing the structure of the output—not a claim of results from a named client.
Six steps from AI learning to approved operating knowledge.
Each stage creates a practical workplace output. The exact sequence is agreed around the selected team and its policies.
Roles and responsibilities
Core responsibilities, recurring work, systems, decisions, dependencies, exceptions and bottlenecks.
Output: Draft role playbooksImportant workflows
Triggers, inputs, steps, decisions, handoffs, approvals, exceptions and current pain points.
Output: Structured workflow mapsAI guardrails
Approved and prohibited uses, information rules, human checks, named accountability and escalation.
Output: Draft AI-use controlsAutomation opportunities
Prioritise by frequency, effort, stability, sensitivity, complexity, benefit and required oversight.
Output: Prioritised opportunity listReview and approve
An authorised manager corrects submissions, resolves conflicts, confirms ownership and approves final assets.
Output: Validated knowledge and controlsUseful assets—not a folder of course notes.
The approved collection can support onboarding, consistency, governance, knowledge retention and future improvement work.
Role playbooks
Responsibilities, recurring activities, systems, decisions and dependencies.
Workflow maps
Reviewed descriptions of how important work moves through the organisation.
AI-use guardrails
Rules showing where AI is allowed, restricted or subject to human review.
Quality examples
Manager-approved examples showing what good work looks like and why.
Knowledge-risk visibility
Processes that depend on individual employees or undocumented experience.
Automation pipeline
Potential AI and software opportunities grounded in real workflows.
Training evidence
Participation, activity, review and completion records from the pilot.
Start where the operational problem is visible.
You do not need to map the whole organisation. Choose one department with a participating manager and a practical reason to act.
- Introducing Microsoft Copilot or ChatGPT Enterprise
- Concerned about unapproved AI tools
- Recruiting or onboarding at pace
- Standardising inconsistent processes
- Preparing for growth or acquisition
- Reducing key-person dependency
- Reviewing operational efficiency
- Exploring workflow automation
- Building an internal AI policy
- Developing departmental AI champions
Choose a team with repeatable, knowledge-heavy work.
Onboarding, account management, communication, review and escalation.
Recurring processes, handoffs, exceptions, approvals and improvement.
Reporting, invoicing, reconciliation and review with strict information controls.
Onboarding, policy access, recruitment administration and employee support.
Research, qualification, proposals, content and approval workflows.
Triage, response, escalation and quality-control processes.
A defined, supported starting point.
- One selected department
- Five to ten participants
- Four-to-six-week guided programme
- Practical, role-specific AI learning
- Role and workflow templates
- AI guardrail activities
- Automation-opportunity assessment
- Manager review and approval workflow
- Pilot progress reporting
- Final leadership review
- Approved employer-owned outputs
The exact scope and output handover are agreed before the programme begins.
A team, a sponsor and time to review.
- One participating department
- A senior sponsor
- An authorised manager
- Five to ten active participants
- Time for workplace capture activities
- Existing AI and information rules, where available
- A final review meeting
TIQ+ provides the programme structure, platform, templates and pilot support.
What this is—and what it is not.
It is
- Practical AI training applied to real work
- A structured way to capture operating knowledge
- A manager-controlled approval process
- A method for identifying responsible AI opportunities
- A focused starting point for wider AI adoption
It is not
- A generic AI-awareness webinar
- An uncontrolled automation project
- A replacement for management review
- A promise that every process should use AI
- A requirement to migrate all your systems
- A consultancy project with no internal ownership
Frequently asked questions
Start small. Keep control.
The pilot is designed to answer practical questions before any wider rollout or automation decision.
Do we need to have selected an AI platform?
No. The pilot can help clarify where AI may be useful and what controls should apply before a wider technology decision is made.
Is this only AI training?
No. Participants develop practical AI capability while creating reusable role, workflow, governance and improvement assets.
Will employee submissions automatically become company policy?
No. Submissions remain drafts until they have been reviewed and approved by an authorised manager.
Who owns the information created during the pilot?
Your organisation retains ownership of its organisation-specific roles, workflows, controls, examples and operating knowledge, subject to the agreed contract and data-processing terms.
Do we need to include the whole company?
No. The recommended starting point is one department with five to ten participants.
Can the pilot work alongside our existing training provider?
Yes. It can complement existing internal training or external provider relationships.
Does the pilot include custom software development?
Not as standard. It identifies and prioritises potential automation opportunities. Any later implementation is scoped separately.
What happens after the pilot?
You can retain and use the approved outputs, extend the approach to another department, maintain the operating-knowledge library or separately explore selected automation opportunities. There is no requirement to expand.
Start with one department
See the two-minute example
Choose a team where the problem is real.
Start where knowledge is difficult to retain, processes are inconsistent or employees are already experimenting with AI. We will help define a focused pilot—not give you a generic platform presentation.
5–10participants
4–6weeks
1department
Discuss a pilot team
Tell us which team and operational problem you have in mind.
Train one team—and keep what they uncover.
See how practical AI learning becomes structured, manager-approved operating knowledge.
The workflow is accurate. Keep confidential data outside unapproved tools, require account-manager verification and use the approved template as the quality standard.