How We Use AI
Effective date: TBD (publish date) Last updated: 2026-05-03
We tell clients up front: we use AI heavily inside Main Street IQ. We think you should know what that means in practice, what it does not mean, and how we keep your data safe while doing it.
If you have questions, email scott@mainstreetiq.com.
What AI does in our work
| Where | What AI does | Human in the loop |
|---|---|---|
| Audits and benchmarks | Drafts findings, synthesizes signals across many sources, scores against rubrics, writes recommendations | Every audit is reviewed and edited by Scott before delivery |
| Research and dossiers | Pulls public data on a company or person, drafts pre-call briefs and prospect dossiers | Reviewed before any external send |
| Content (blog, LinkedIn, email drafts) | Drafts in MSIQ voice from a brief Scott writes | Scott reads, edits, and signs every external piece |
| Operational analysis (cash flow, financial commentary) | Drafts narrative around your numbers | Always reviewed; the analytical judgment is human |
| CRM and pipeline | Categorizes leads, drafts replies, builds briefings | Scott reviews and sends |
What AI does NOT do
- It does not make final recommendations to your business without a human reviewing them
- It does not send emails to you or your customers without Scott's review
- It does not autonomously execute financial transactions, sign contracts, or change your systems
- It is not used as a substitute for licensed advice (legal, tax, financial regulatory)
A human is responsible for every deliverable that leaves Main Street IQ. Scott Hess. Names attached. Edits visible. Errors are ours, not the model's.
What this means for your data
When we use an AI tool, your information may pass through that tool's API. We use the API tiers of the providers below, which means:
- Your prompts and outputs are not used to train the providers' models
- Providers retain the data only briefly for abuse monitoring (typically 30 days or less) and then delete it
- Data flows are encrypted in transit
- Providers we use: Anthropic (Claude), OpenAI (GPT-4 family), Google (Gemini), Perplexity
We do not feed regulated personal information into AI tools. Specifically: - No Protected Health Information (PHI under HIPAA) - No payment card numbers or full bank account credentials - No Social Security numbers or government IDs
If your engagement requires a stricter posture (HIPAA, financial services, defense), we will scope around the affected data or decline the work.
For complete sub-processor details, see our Privacy Policy or request the full list at privacy@mainstreetiq.com.
Why we built it this way
Most agencies have started using AI in the last two years. We built MSIQ AI-native from day one because that is the only way one person can deliver the depth of work this practice demands at the price point our clients can afford. The leverage is real. The accountability is also real, and it stays with a human.
If you ever want to know whether something we delivered was AI-assisted, the answer is yes. If you want to know whether a human stood behind it before it reached you, the answer is also yes.
Changes
When we add or remove AI tools from the stack, we update this page and our Privacy Policy sub-processor list. For active clients, material changes are emailed at least 30 days in advance.
Contact
scott@mainstreetiq.com Main Street IQ LLC, San Luis Obispo, California