# AI Team Workshop Training Packet

Sample buyer-facing packet generated from Setup Your AI private product repositories.

Source: setupyour-workshop-products
Data policy: sample fixture only; no live customer records, credentials, uploads, payment data, or private transcripts.
Approval policy: human approval is required before external sends, customer-system writes, booking holds, pricing exceptions, payment-affecting actions, legal answers, refunds, or policy exceptions.

# AI Workshop Plan: Mountain View Events

Operator: Jim
Format: remote half-day
Audience: Owner, venue coordinator, and two sales assistants

## Readiness

- 3.88/5 - customize_first

## Team goals

- Use AI to answer common tour and availability questions faster
- Create safer prompts for follow-up drafts and internal summaries
- Find the first workflow automation candidate after training

## Current tools

- Gmail, Google Calendar, Squarespace, Google Drive

## Agenda modules

1. AI foundations for real work: Team understands useful AI patterns, limits, and review habits.
2. Approved facts and answer boundaries: Team separates approved business facts from guesses and exceptions.
3. Prompt patterns and reusable drafts: Team leaves with reusable prompts for drafts, summaries, and checks.
4. Workflow discovery and ROI triage: Team identifies the first workflow worth automating after training.
5. Agent operations and human approval: Team knows where AI can draft, where people approve, and when to escalate.

## Exercises

- AI foundations for real work: Rewrite one real internal request into a clear AI task with context and constraints. Success check: Each participant can explain what the AI should and should not do.
- Approved facts and answer boundaries: Build a mini approved-facts sheet for common customer questions. Success check: Risky questions are flagged for human review instead of guessed.
- Prompt patterns and reusable drafts: Turn a messy customer request into a structured reply draft and review checklist. Success check: Draft includes missing details, risk flags, and next action.
- Workflow discovery and ROI triage: Score three repeated workflows by value, repeatability, data readiness, and risk. Success check: One first-build candidate is selected and one no-build call is named.
- Agent operations and human approval: Classify sample customer messages into safe draft, needs owner approval, or stop. Success check: External-send and customer-system-write boundaries are explicit.

## Guardrails

- Use approved facts and customer-provided context; do not invent policies, prices, availability, or commitments.
- A human approves external customer messages, customer-system writes, payment-affecting decisions, legal answers, refunds, and policy exceptions.
- No private customer data, credentials, or sensitive transcripts are copied into tools that are not approved for that data.
- Escalate these boundaries immediately: pricing exceptions, legal questions, refunds, contract changes.

## Desired outcomes

- Team can draft safe replies from approved facts
- Owner gets a prioritized follow-up build recommendation
- Everyone leaves with three reusable prompts

## Follow-up actions

- Jim: Send the workshop recap, reusable prompts, and risk-boundary checklist. Proof: Recap packet names prompts, guardrails, and selected first-build candidate. Window: 14 days.
- Mountain View Events: Pick one team habit and one workflow candidate to test for two weeks. Proof: Team names the owner, representative examples, and success metric. Window: 14 days.
- Jim: Recommend whether the next paid step is an Opportunity Map, automation sprint, or agent build. Proof: Follow-up recommendation includes price anchor and non-goals. Window: 14 days.

## Non-goals

- No production agent launch during the workshop.
- No autonomous external sending or customer-system write access.
- No CRM, calendar, email, payment, or website integration unless separately scoped.

## Pricing anchor

- Foundations workshops start around $2,500 remote.