Your team is already using AI
Not next year. Today. Staff across New York businesses are pasting client files, designs, financials and contracts into consumer AI accounts, usually with good intentions and almost always without anyone approving it. Unless the training setting is switched off, those accounts may use what’s pasted in to improve their models. Either way, the data has left your control.
That’s Shadow AI, and most firms discover it by accident. Not through a breach, usually, but through a conversation: someone mentions the tool they’ve been using for months and the room goes quiet.
Haber Group works differently. We find out what’s actually in use, secure it, then help your team use AI properly. If you already have an IT provider you’re happy with, we’ll work alongside them. You don’t have to switch anyone out to get this fixed.
THE PROBLEMS WE SOLVE
Six things that come up in almost every conversation we have about AI with New York businesses. Most firms have at least four of them.
Shadow AI
Staff are using consumer AI accounts with company and client data, and nobody has a list of what’s in use. You can’t govern what you can’t see, so we start by finding out: which tools, on whose accounts, with what data.
No AI policy
Staff are using consumer AI accounts with company and client data, and nobody has a list of what’s in use. You can’t govern what you can’t see, so we start by finding out: which tools, on whose accounts, with what data.
Client and contractual risk
Your clients are starting to ask how you handle AI, and some contracts already prohibit putting their data into third-party tools. We help you answer honestly, and close the gaps before the question arrives.
Data leaving your control
Once information is in an unsanctioned account, you have no record of it and no way to retrieve it. We put controls in place that stop company and client data reaching platforms you haven’t approved.
Low adoption of approved tools
Some firms buy licenses and watch them go unused, while staff quietly carry on with the tools they already know. Training and real use cases matter more than the license itself.
Wasted hours nobody has mapped
People use AI for emails and meeting notes while the repetitive work that actually costs hours goes untouched, because nobody has sat down and mapped it. That’s where the return is.
HOW HABER GROUP HELPS
Secure it. Use it. Automate it.
AI isn’t one project. It’s three, and they have to happen in order. Securing what’s already in use comes first, because adoption on top of an unsecured environment just multiplies the exposure. Then you get people using approved tools properly. Then you find the repetitive work worth automating.
We can start anywhere in that sequence, but we’ll tell you honestly if you’re trying to skip a step.
- Find out which AI tools are in use and on what accounts
- Write an AI policy your team will follow
- Put controls in place around company and client data
- Set up one approved platform securely, with training off
- Train everyone, then train the people who’ll go deeper
- Map the workflows worth automating, and build the automations
THE HABER GROUP AI PATHWAY
Three phases, in the order that actually works. Each phase stands on its own and leads into the next. Most firms start at the beginning, because that’s where the risk is.
Secure Start
An AI use policy, controls that stop data reaching unsanctioned platforms, one approved AI tool set up properly with single sign-on and training switched off, plus AI awareness and prompt training for everyone. A fixed-scope project with a clear end.
Managed AI
AI tools and risks change every month, so a one-time lockdown drifts out of date. We manage licenses, keep controls and settings current, maintain the policy, support your users, and run a quarterly check on what’s being used and what’s been blocked.
AI at Work
Having a secure AI tool is not the same as people using it. We set AI up inside the tools your team already lives in, build use cases by role, and train people on when to use it and when not to. Adoption, not licenses.
HOW WE'RE DIFFERENT ON AI
Most IT providers in NYC haven’t raised AI with their clients at all. Here’s where that difference shows up.
Risk first
Most AI conversations start with productivity. We start with what's already leaving your business, because that's the problem you have today rather than the one you might have next year.
Alongside your IT
Most AI conversations start with productivity. We start with what's already leaving your business, because that's the problem you have today rather than the one you might have next year.
Policy that sticks
You don't have to change IT provider to get your AI governed. We work hand in glove with your existing provider. Instead of your IT partner, we'll be your AI partner.
Platform agnostic
We set up the platform that fits your business rather than the one we happen to resell, configured properly with single sign-on and model training switched off.
Built in New York
We've supported New York businesses since 2000, including firms where client confidentiality is the whole business. We know what a breach of it actually costs here.
Accountability
We share monthly SLA reports with clients: 96% response, 97% resolution. The same accountability applies to AI work as to everything else we do.
Common questions about AI governance
Is it safe to let staff use ChatGPT at work?
It can be, with the right setup. The risk isn’t the tool, it’s the account. A personal consumer account may use what’s pasted into it to train the model unless that setting is switched off, and either way the data has left your control. A properly configured business tenant with single sign-on is a different proposition entirely.
What is Shadow AI?
Any AI tool your team uses that the business hasn’t approved or doesn’t know about. It’s rarely malicious. Someone finds a tool that helps, starts using it, and tells a colleague. Six months later it’s part of how work gets done and nobody in leadership has heard of it.
We’re happy with our current IT provider. Does that rule you out?
No, and it’s one of the more common situations we work in. We’ll work alongside your existing provider on AI specifically. You keep the relationship you have and get the part they haven’t covered.
Do we need an AI policy if we’re a small team?
Especially if you’re a small team, because informal rules stop working the moment someone new joins or a client asks the question formally. A policy for a ten-person firm can be two pages. It just has to exist and be signed off.
Where are you based?
530 Seventh Avenue in Midtown Manhattan, which means on-site support across NYC when remote support isn’t enough.
Do you support both Mac and PC environments?
Yes. Mixed Mac and PC environments are normal in design studios, and we support both, including the licensing, file sharing and network complications that come with running them side by side.
Start with a conversation about what’s already in use.
Three starting points, all free and all useful whether you hire us or not. Pick the one that fits where you are right now.
Request a Shadow AI Assessment
One question, answered fast: what AI is your team already using, on whose accounts, and with what data? You get a findings report & a prioritized list. If you want the fuller picture, the readiness assessment goes further. 45 minutes.
Get your free AI policy template
A plain-English AI acceptable use policy you can adapt in an afternoon. Which tools are allowed, for what, and with which data. No legalese, no fluff, and nothing your team will need a lawyer to interpret.
15 minutes.
Request an AI Readiness Assessment
A full review of where your business stands on AI: what's in use, where your sensitive data sits, how ready your security setup is, and where AI could help first. You get a findings report with a prioritized list of what to fix. 60 minutes.