Your staff are probably using AI already: a chatbot in a browser tab, a meeting recorder somebody signed up for, an assistant built into a tool you already pay for. The real question is whether they’re using the right tools, with the right settings, on the right data. We help Central Florida businesses pick the AI systems that fit, set them up properly, write the rules, and keep them managed after launch.
Who this is for
- Businesses whose people already use AI without a policy, and want to know what’s going where before something ends up in the wrong place.
- Firms with rules to follow: law, accounting, medical, financial. Client confidentiality, HIPAA or the FTC Safeguards Rule still apply when the work goes through an AI tool.
- Owners deciding whether to buy Microsoft 365 Copilot or another AI plan, who want a straight answer on whether it’s worth it for their office.
- In-house I.T. teams who have been handed “do something about AI” on top of everything else, run co-managed: you keep the keys and the roadmap.
What we do
- Find out what’s already in use. Which AI tools your staff are signed up for, which apps have been given access to your Microsoft 365 mail and files, and what data is going into them.
- Help you choose. Microsoft 365 Copilot, ChatGPT, Claude and Google Gemini all sell business plans, and they aren’t interchangeable. We match them to the work your people actually do and the systems you already run, and compare what each vendor’s business terms say about retention and training on your data. Sometimes the answer is “not yet”.
- Clean up access first. An AI assistant shows each person whatever they can already open. Old “anyone with the link” shares and over-broad folder permissions turn into answers. We find and fix those before the assistant goes live.
- Compliance setup. A written AI acceptable-use policy that fits on a page or two, data-handling rules for confidential and regulated information, sign-in with multi-factor authentication, logging and retention settings, and data-loss-prevention rules where your licences include them.
- Rollout and training. A pilot group first, then everyone, with short training on what to use AI for, what never to paste into it, and how to check its work.
- Ongoing management. Licences and seat counts, new features as vendors switch them on, access reviews, security settings against new attacks such as hidden instructions planted in emails and documents, and policy updates when the rules or the tools change.
What we don’t do
- We don’t give legal advice. We’ll set up the controls and write the policy in plain English. Your attorney signs off on anything that turns on the law.
- We don’t push one vendor. We recommend what fits your work and your budget, and we’ll tell you when a tool you’re paying for isn’t earning its keep.
Who does what
| Task | SquareWave | You | The AI vendor |
|---|
| Choosing tools and plans | Recommends, with the reasons | Decides | — |
|---|
| Permissions and data clean-up | Does it | Says who should see what | — |
|---|
| AI policy | Drafts it | Approves it (with your attorney where needed) | — |
|---|
| Security and compliance settings | Sets up and maintains them | Approves changes | Provides the controls |
|---|
| Training staff | Runs it | Makes time for it | — |
|---|
| The AI model itself | Watches for changes that affect you | — | Runs and updates it |
|---|
What it looks like in practice
- A manufacturer rolling out Claude. We set up the business plan with Microsoft sign-in, so access follows a group in their directory rather than a list of passwords. Then we connected it to their ERP data through a gateway inside their own Azure network: read-only, open only to the people in the right groups, and with every query logged. Before we called it done, we ran a security review and fixed what it found.
- A property-management group with rules about ChatGPT. Their web filtering blocks unapproved AI sites by default. ChatGPT is opened up one person at a time, and a named manager approves each one.
- Our own shop. Our technicians use AI connected to our ticketing, monitoring and knowledge base, through our own gateway with Microsoft sign-in and access granted per person. Those are the same controls we set up for clients, and we’ve run into the rough edges ourselves first.
Questions people ask
Is it safe to put client information into ChatGPT?
It depends on the plan, the settings and the client agreement, not the brand. A free personal account and a managed business plan are very different things. We’ll check what your people use and what your obligations allow.
Do we need Microsoft 365 Copilot?
Maybe not. If your work lives in Outlook, Teams, Word and Excel it can save real time, but only after the permissions clean-up. If not, a different plan, or none, may fit better. We’ll tell you which, and why.
We’re small. Do we really need an AI policy?
Yes, and it can be short. Your staff are deciding what to paste into AI tools every day. A page of clear rules beats everyone guessing.
Can you work alongside our I.T. person?
Yes. That’s the co-managed model.
What do you need from us to start?
Admin access to Microsoft 365 or Google Workspace, a list of the AI tools you know about, and 30 minutes with whoever runs the office.
Talk to us about AI at your office or call (407) 204-9874. Related: all services.