Most LOs know AI is happening but haven’t started using it because nobody’s told them, in plain English, what’s safe to type into a chatbot. Here are three tools you can have running by Friday: what each one is for, the LO use case, and what to never put in it.
One disclaimer: I’m not a compliance officer and this isn’t legal advice. Run any AI workflow that touches borrower data past your compliance team before you scale it.
The compliance frame, in one paragraph
Free tiers of ChatGPT, Claude, and Gemini may train on or retain user inputs depending on the product and the day’s privacy policy. Borrower names, socials, addresses, and income figures don’t belong in a free public chatbot. The fix isn’t “don’t use AI.” The fix is to use the right tier (Teams, Enterprise, or API-with-zero-retention) for anything client-specific, and use free tools for generic drafting and research. That distinction is the whole game.
Tool 1: Claude (Anthropic): for writing
What it does well: Email drafting. Plain-English rewrites. Long-doc summaries. Tone control.
| Plan | Retention | Notes |
|---|---|---|
| Free / Pro | Inputs may be used per current terms | Drafting only. No client data. |
| Teams | Zero training on inputs | Safer for internal workflows |
| API (zero-retention) | Enterprise-grade | What custom HTS builds run on |
LO use case: the borrower follow-up rewrite. Two awkward paragraphs become a tight, on-tone email in under a minute:
Model: Claude (claude-opus-4)
Prompt: "Rewrite this email in a warmer, more direct tone.
Keep it under 120 words. End with a single clear ask.
Don't change any factual details."
What to NEVER put in the free tier: borrower names, SSNs, DOBs, account numbers, pay stub or W-2 figures, bank statements, full email threads with PII. Strip the PII out, get the rewrite, paste the names back in. That’s the workflow.
Drafting in AI and editing in your CRM is faster than drafting in your CRM. Treat the AI like a writer, not a database.
Tool 2: Perplexity: for market intel and prospect research
What it does well: Web-grounded research with citations. Faster than Google for any question that needs a synthesized answer.
LO use case: the Monday-morning market update.
"What were the major US mortgage rate movements last week?
Cite Mortgage News Daily and the MBA Weekly Applications Survey."
You get a synthesized answer with links in 90 seconds. Paste highlights into a market-update email, add your commentary, send to your top 30 past clients and realtor partners.
What to NEVER put in it: anything from a borrower file, internal pipeline data, anything covered by your shop’s NDA. Perplexity is a search tool with the same posture as Google: smarter results, same boundaries.
Tool 3: An AI notetaker: for meetings
What it does well: Joins your Zoom, Teams, or Google Meet calls, transcribes them, produces a summary and action items.
Fathom, Otter, and Fireflies do roughly the same job. Pick the one that matches your stack and your shop’s data agreement. The point is to have one running on every realtor lunch, borrower discovery call, and internal meeting you’d otherwise forget by Friday.
LO use case: the realtor partner meeting. Notetaker joins, transcribes, produces a four-bullet summary plus action items. Paste action items into Jungo. Set a two-week follow-up task. Done in 90 seconds post-call. The same workflow without it is 15 minutes and frequently doesn’t happen.
What to NEVER do: record without verbal consent (the notetaker should announce itself, or you should). Don’t auto-share full transcripts of borrower calls externally: summaries yes, transcripts no. Don’t send internal pricing or pipeline transcripts outside your shop.
Most of an LO’s intel about referral partners lives in the conversations themselves. If those conversations aren’t captured, the database can’t help you remember.
The 30-minute setup
- Claude Pro: drafting, rewrites, summaries. PII stays out.
- Perplexity Pro: Monday market updates and realtor research.
- An AI notetaker: on every meeting.
Total: under $70/month. Setup: about 30 minutes. Time saved per week, conservatively, in HTS engagements I’ve watched: 3 to 5 hours.
What this is not
This is the starter stack: three off-the-shelf tools any LO can install today without IT support. It’s not a custom workflow or an AI agent running your business.
If you want AI plugged into Jungo, your email, your phone system, and your loan pipeline (end-to-end), that’s the next conversation. We build that inside the HTS Operating System tier. We did a version inside a recent HTS engagement and the most-used piece was a 90-second daily summary, not the fancy parts.
Ready to go further?
If the starter stack is in place and you want a real workflow built around it, book a discovery call. The call is 30 minutes. You’ll leave with a one-page plan whether or not you hire us.
Compliance note: This post discusses AI tools in the context of mortgage workflows. None of the recommendations here is a substitute for review by your shop’s compliance officer. Any workflow that processes borrower PII should be vetted before deployment. Pricing and plan features for third-party tools change frequently: verify current terms with the vendor.
