要点
An internal AI day works best when it has one shared objective and different routes for different participants. Combine an opening context session, a grounded example, optional practical breakouts, structured questions and a closing next-step record. Confirm tool access, staffing, accessibility and production responsibilities before running exercises. The sample agenda is adaptable and is not a booked AISB event.
An internal AI day can create useful energy, but a full room does not tell you what people can do the following week. Our starting point is to decide what the audience should understand, practise or choose, then build the programme around that outcome.
Choose an objective and audience
Choose one primary objective: create a common understanding, practise a defined task, identify workflow opportunities or engage a client audience. Supporting objectives can fit around it, but they should not compete for the same time.
Segment participants by role and starting knowledge. A finance manager, a developer and a customer-facing employee may need different examples and review responsibilities. Ask participants about tool access and experience before finalising practical tracks; do not treat an “AI beginner” label as a complete learning need.
For a client event, remove internal-only material and clarify what participants can share publicly. For an employee day, make approved usage rules and internal support easy to find.
Sample one-day agenda
Illustrative agenda. Breakout capacity, timing, contributors and equipment need confirmation for the actual audience.
| Time | Session | Purpose | Output / owner |
|---|---|---|---|
| 09:00–09:30 | Sponsor opening and audience questions | Explain why the day matters | Shared objective and questions / internal sponsor |
| 09:30–10:15 | Practical AI context | Establish a common vocabulary and limits | Terms and misconceptions / session lead |
| 10:15–10:45 | Workflow example and discussion | Show inputs, output, review and limitations | Questions worth investigating / practitioner |
| 11:00–12:00 | Parallel practical tracks | Match activities to starting skills | Reviewed exercise or workflow map / track leads |
| 12:00–12:30 | Lessons from the tracks | Identify what worked and what failed | Evidence and unresolved issues / rapporteurs |
| 13:30–14:30 | Business challenge breakouts | Turn problems into candidate experiments | Short use-case cards / business owners |
| 14:30–15:15 | Tools, data and governance questions | Surface constraints before experimentation | Dependencies and escalation route / internal specialists |
| 15:30–16:15 | Next-step clinic | Agree what should happen after the day | Owners and review dates / sponsor |
| 16:15–16:30 | Closing and feedback | Explain follow-up and collect useful feedback | Feedback and next communication / organiser |
Leave breaks and transition time in the actual running order. If the group cannot split into supported tracks, use a demonstration with participant analysis rather than an unsupported hands-on exercise.
Design practical tracks
One beginner activity can ask participants to review a fictional AI-generated response: what is missing, what is unverifiable, and what needs a human decision? A business track can map a workflow and propose a bounded experiment. A technical track can test an integration in a permitted sandbox.
These are example activity designs, not exercises validated on a client cohort. Before delivery, test the selected task on the actual tools, check permissions and identify an accessible alternative. Use synthetic or specifically approved materials. Never ask participants to upload confidential files simply to make a demonstration realistic.
The use-case prioritisation template gives business breakouts a consistent output. The format comparison helps decide whether a smaller workshop or training session would be more useful.
Agree production responsibilities
The proposal should identify who owns the programme, contributor contracts, venue, AV, connectivity, participant communications, accessibility, exercise support and contingency decisions. An editorial programme lead does not automatically own venue production or internal tool approval.
| Dependency | Confirm before delivery |
|---|---|
| Contributors | Topic, current affiliation, availability, fees and recording terms |
| Practical exercises | Accounts, access, permitted data, licenses and support |
| Breakout rooms | Capacity, accessibility, facilitators and transitions |
| AV and connectivity | Equipment, tests, on-site contact and fallback |
| Recording and photography | Permission, distribution and participant information |
| Follow-up | Owner, material, review date and separate delivery scope |
The provider brief and scope framework make those responsibilities easier to compare across proposals.
Measure what happens afterwards
Separate attendance, satisfaction, learning and continued use. Each answers a different question. A short feedback form can tell you which sessions were useful; an assessed task can show what participants can do; a later check can show whether the approved workflow is still used.
Do not describe a positive reaction as productivity improvement. Record a starting point and include non-responses when reporting follow-up. Use the thirty-day plan to assign actions after the event.
Methodology
About this agenda
AISB editorial planning template, published on 4 October 2026. Timings and activities are illustrative. No client results, fixed capacities or contributor bookings are implied.
Scope your company AI event
Tell us who the event is for, what they should take away and which production responsibilities need support.
If this interests you
- Private AI events: review the AISB programme and production scope.
- Executive briefing: choose a smaller format for one leadership question.
- Thirty-day follow-up plan: turn the day’s outputs into owned next steps.
Continue the discussion
Explore a company AI programme. Connect this reading with the published Summit programme and available recordings.
常见问题
What should an internal AI day include?
A clear purpose, relevant context, practical examples, appropriate audience tracks, time for questions and named next actions. Practical exercises also require approved tools, permitted data and enough support for the group.
How do we include beginners and experienced users?
Use a shared opening, then offer different tracks with explicit prerequisites. Beginners can analyse a workflow or practise output review; experienced participants can test a more specific task using approved tools.
Does an internal AI day replace staff training?
No. It can establish context and identify needs, but skill development needs practice, feedback and follow-up. Attendance or satisfaction alone does not demonstrate competence.
Change log
- — Shortened search metadata, aligned the named author and added contextual learning links; removed statistical callouts.
关于作者
- TWTanguy WinckerCo-founder, AI Summit BarcelonaTanguy is a co-founder of AI Summit Barcelona. His editorial work covers company programmes, event planning and the Barcelona ecosystem.
