Speaking & Training
AI keynotes, workshops,
and team training
Some groups haven’t touched AI yet and are trying to figure out whether it’s worth doing as a team. Others already tried it and couldn’t turn it into repeatable work. Either way, start with the problem. I’ll help you pick the format.
Keynotes
A keynote gives a room one useful way to think about AI, before everyone scatters and starts trying five different tools on their own. It works for conferences, associations, leadership meetings, and company all-hands, anywhere the first need is a shared starting point, not a full workshop.
Interactive workshops
A workshop takes one real problem and pushes it far enough that people actually build something in the room. These are starting points, not fixed scripts. I change the examples and exercises to match what your group actually does all day.
AI 101 for Teams: What AI Can Do and How to Start
We cover what generative AI actually does, how chat and work tools differ, what context changes the answer, which information needs extra care, and how to choose one sensible first job. People leave with their own written instructions for the AI they use, plus a real plan for one experiment.1
I’ve taught this exact format live, for a staff team with mixed roles and very different comfort levels with AI.
How to Stop Getting Generic AI Answers
We look at what the AI can currently see, what it keeps between conversations, which source material belongs with the job, how examples teach the standard, and how to correct a bad answer without starting over. People build a reusable context brief for one kind of work, and a request-and-check template they can use again.
How to Find Work AI Can Actually Take Off Your Plate
We map one repetitive job: what starts it, what information comes in, which steps repeat, where judgment belongs, what can go wrong, and how the team knows the job is done. People leave with a map of the current process and a small, supervised test that can prove whether AI actually helps before anyone builds more.
How to Use AI Without Losing Your Voice
We pull real examples of the organization’s voice, separate observed patterns from assumptions, name the facts and boundaries the AI can’t invent, and test the guide on work the team already produces. People build the first version of a reusable voice guide, then test it on one real piece of writing.
A living guide for the session and the work afterward
Selected workshops and training series come with a branded interactive guide built for that session. It holds exercises, examples, copy-ready prompts, reflection questions, and space for the work people create.1
After the session, the same guide gets updated with the recording, timestamps, answers from the room, examples shaped to your organization, and next steps. Nobody has to dig up the slides and a separate workbook three weeks later.1
The guide is its own demonstration. People see exactly what AI can help build when someone gives it real context, checks the work, and builds for a specific group instead of asking it to “make something cool.
View a Sample Interactive Workshop Guide
Team training that builds one working system
4-part series: Build a Working AI System for Your Team
The workshop guide grows with the series. Each session adds the decisions and working materials the next one needs, so the final guide becomes the team’s home for the pilot, not another training handout.
Session 1: Understand the tools and set the boundaries
The team learns what the tools can access, which jobs fit each type of AI, what information needs protection, and where a human has to stay responsible. By the end, the team has a short AI use guide and a map matching current tools to current jobs.
Session 2: Give AI the business context it keeps missing
The team decides which facts, standards, examples, and source files an AI needs before it can do useful work, and keeps changing information separate from durable information so the setup doesn’t go stale the minute the business changes something. That becomes a shared context pack for the pilot area.
Session 3: Choose and map the first workflow
The team looks at repetitive work, access limits, sensitive information, failure points, and the parts that still need judgment, then picks one job small enough to test and useful enough to matter. The team leaves with a ranked list of possible jobs and a full map of the one it picked.
Session 4: Build, test, and assign the working pilot
The team runs the selected workflow with real inputs, checks the result against a defined standard, fixes the weak spots, and decides who owns it after training. The final session produces one tested pilot, with its instructions, approval points, completion check, and an owner.
The series can stop there. More tools, workflows, or automation only make sense once the first pilot actually works.
Other team
training options
• When nobody’s started yet, and you’re trying to decide if this is an individual thing or a team thing.
• When AI use is already uneven across the team.1
• When managers don’t know how to review AI-assisted work.1
• When everyone has their own prompt folder and nothing shared.1
• When nobody owns the system after training ends.
AI Consulting
Consulting is for leaders who know AI could help but don’t know which job deserves attention first, what the tools need access to, or how much autonomy is actually safe. It can stop at a clear plan, or go all the way through a working build.
I can help you review the work people repeat and where it currently gets stuck, choose a short list of AI jobs worth testing, identify missing context, access, privacy, and approval decisions, scope one pilot with an owner and a clear definition of done, and review the result before the company expands it.
Choose the level of help you need:
Strategy and system design. I find the right starting point, map the workflow, define the guardrails, and hand your team a build plan with clear acceptance criteria.
Supervised implementation. Your in-house team or existing vendor does the build. I direct the work, review milestones, test the finished workflow, and keep it tied to the actual business problem.
Implementation with my contractors. I scope the system and supervise a contractor from my network through the build and testing. The proposal makes it clear who’s responsible for each part before work starts.
Who these sessions are for.
Companies and small business teams. Conferences and professional associations. Leadership and management groups. Entrepreneur and member communities. Teams starting with generative AI. Teams that tried AI but haven’t turned it into dependable work.
Not sure? Tell me who’s in the room, where they’re stuck, and what you want them walking away with. I’ll tell you whether that’s a keynote, a workshop, a series, or a consulting project.
Tell me where you're starting. I'll look at it myself and we'll have a short call. No deck. No obligation.
I'm Kay Phillips, an entrepreneur, a practical AI educator, and a woman with ADHD, running a business from a 40-acre farm outside Denver.
I’ve built two six-figure businesses through relationships and referrals, mostly through stubbornness and figuring things out the hard way. I also built a pretty impressive planner graveyard trying to make systems designed for somebody else’s brain work for mine.
So I stopped treating my brain like the broken part of the process.
I got serious about AI because my business depended on me remembering too much. Every tool knew a different version of the business. A new chat meant another round of explaining who I was, what I was building, and which decision had replaced the one hiding in an old file.
That’s why I built KayOS.
It gives the AI tools I use one current business record, clear rules, and a way to verify the work, instead of reporting success because a document exists somewhere.
And I teach from the parts that work after the clean demo is over. A system that only works on a perfect Tuesday is a pretty bad system.
How I approach the work.
Tell me where you're starting. I'll look at it myself and we'll have a short call. No deck. No obligation.