HIPAA and AI documents, note-writing tools, and a cold email fix - Connecting AI like ChatGPT or Claude to your documents can be HIPAA safe if you set it up right. There is no perfect fix except a BAA, and even that has gaps. - Most systems built in ABA back office skip HIPAA entirely. They run on data and binary numbers instead of client names. - For systems that do touch client data, a script strips client names out before anything goes to Anthropic, then subs them back in after. - Marketing automations, scheduling, and email scans can all run without touching HIPAA data at all. - For notes and reports, most EHRs already include AI note writing. Frontera got a good mention for report writing. Introverbal AI has an assessment tool too. - Assessment writing mainly needs a transcript, data on what the kid does and doesn't do, and BCBA input. All three are easy to gather now. - No solid CEU exists yet on AI workflows in ABA. The Shaping AI conference on Behavioral Live was a good one, though it may not be active anymore. - A cold email campaign stalled in week two. Checking three known failure points instead of asking Claude to diagnose it found the bug fast. The automation was only uploading part of the variable set, so emails weren't personalizing right. Watch today's full video: https://www.skool.com/the-aba-back-office/classroom/ad8c34bd?md=5f1e661d80ac41cfa58f52bd2692d421

ABA Back Office Daily Recap: HIPAA Rules, AI Scheduling, and EHR Notes - This is the first daily video for the ABA Back Office community. New videos are planned Monday through Friday, mixing Q&A with a topic discussion. - The ABA Back Office plugin helps behavioral operators use AI for real reports, findings, and automations on their own computer, beyond a basic chatbot. - Automations split into two groups: protecting money (projections, underpayment audits, watchdogs) and running or growing the clinic (large projects like speed to lead or cold email). - HIPAA rule: Anthropic will not sign a business associate agreement. Client data has to stay local, moving only from one file to another on your own computer. - Minimize client names wherever possible. A built-in substituter script strips names locally before data leaves your computer, getting you about 95% of the way there. - AI works best on daily or weekly tasks with minimal decision making, like scheduling. Feed it the data and rules, then let it test many schedule options and pick the best one against a KPI. - On EHR AI notes: Central Reach's version gets decent feedback but costs more. Most EHRs will have this feature soon, so it's a weak reason to pick one platform over another. - On AI compliance audits like Brellium: the right setup depends on note volume and how well the audit fits into daily workflow without extra manual steps. - On AI use in interviews: he hasn't seen it happen and thinks it would be obvious. Digging into details beats switching to in-person interviews.

Check out this video to see how affiliates work for The ABA Back Office! 1) Look at your profile in the top right 2) Find The ABA Back Office in your affiliate section 3) Give that link out and get paid 40% of their monthly subscription cost EVERY month they renew TLDR- Refer 3 people, get paid to be a part of The ABA Back Office!

Aiming to have the next large automation project built out by September 2026- tell me what you want me to build! To be clear, we have PLENTY of skills and automations already in the group, and we will still be putting out new skills every few weeks. The large AI builds are more complex installs, often involving webhooks, APIs, etc, so it needs more instructions and build-out.

Before you run anything with client information, read this. It is the most important post in here. I am a BCBA. I know how it feels to be handed an AI tool and wonder if you just broke HIPAA. So I built this whole thing around one rule. Your client data stays on your computer, and the AI sees as little of it as possible. I know this post is long, but its important. HIPAA violations are no joke, and the AI companies don't sign BAAs. Your data stays local. Everything runs through Claude Code on your own machine. The systems open your files right there, do the math or the writing, and show you the result. Your files are not uploaded to me, to a server, or to anyone else. There is no cloud database of your clients sitting somewhere. Most systems never touch a client name. The money and billing systems, like projections, underpayment, auth and cancellation tracking, and scheduling, read your export on your computer and pull out totals and patterns. The report you get back has no client names in it. Just payers, codes, dates, and dollars. When a system needs a specific client or claim number so you can act on it, that goes into a worklist file that stays on your machine, and the AI does not read those rows. A plain script writes them straight to disk. The one system that reads a real document protects it first. The denial and appeal writer has to read the actual letter. Before it does anything, it swaps every name, date of birth, member ID, and claim number for a placeholder. It writes the appeal over the placeholders. Then it puts your real values back on your computer and checks the result to make sure nothing slipped through. Every system tells you its setup before you run it. You always know what a system does with your data before you hand it a file. No surprises. About Claude and your inputs. Whether Anthropic uses what you type depends on your Claude plan and your settings. It is not the same on every tier, so check your plan's data settings. And remember the design already keeps client data out of the AI's hands as much as possible. Most of what you run never puts a client record in front of the model at all.