How We Handle Your Data

This page explains, in plain language, how Tri-Cities AI Lab thinks about data when working on a project, and what you should and shouldn't send us through a standard website form. It's not a substitute for a signed engagement letter or contract, and it doesn't create legal guarantees this business can't back up. If your project has specific data-handling requirements (a client agreement, a regulatory obligation, a court order), those govern, and we'll work within them.

A practical approach to client data

There's no single safe way to handle every kind of information, because the right approach depends on the situation, not a one-size-fits-all policy. What actually matters for a given project:

• How sensitive the data itself is
• What your organization already requires, contractually or internally
• Which tools and platforms make sense for that sensitivity level
• How the work gets deployed and who else can access it
• Any contractual obligations already in place with your own clients or partners
• Legal or professional responsibilities that apply to your industry (legal, medical, financial, and similar fields all have their own rules)

We figure this out at the start of a project, not after something has already gone wrong.

What you should not send through ordinary forms

The contact form, the AI Readiness form, and the Start Here questionnaire on this website are standard web forms. They are not a secure file transfer system, an encrypted vault, or a HIPAA-compliant intake portal. Please do not submit any of the following through those forms:

• Passwords or login credentials
• Social Security numbers
• Complete financial account information (full card or account numbers, routing numbers)
• Medical records or protected health information
• Privileged legal information or attorney work product
• Confidential investigative information
• Other protected personal data
• Trade secrets
• Highly sensitive client records

If a project involves any of this, tell us that in general terms through the form (for example, "we handle patient records and need to discuss how that affects the build") and we'll arrange a more appropriate way to share the actual details, whether that's a call, a secure document exchange, or something else that fits the situation.

How project information may be handled

Once a project is underway, the general principles we work from:

• Project information is limited to what's reasonably needed to do the work, not collected just because it might be useful someday
• Access to client information is limited to the people who actually need it for the project
• Tool selection depends on the sensitivity of the work. A general-purpose AI chat tool is not treated the same as a system built for regulated or confidential data
• Enterprise or private deployment options are recommended when the sensitivity of the work calls for it
• Client data should not be pasted into consumer-grade AI tools casually, especially anything that could train a public model on it
• Model training and data retention settings on any AI platform we use get reviewed against what a project actually requires, not accepted as the default
• Human review stays part of the deliverable. AI-assisted work gets checked by a person before it's called finished
• When a project genuinely requires a more confidential transfer method than email or a web form, one can be arranged

Third-party services

This website and client project work may involve third-party providers for things like website hosting, scheduling, email delivery, analytics, file transfer, AI platforms, and automation tools. Which providers make sense for a given project is evaluated case by case, based on what that project actually needs, not a fixed list applied the same way to everyone. See the Privacy Policy for the specific third-party services this website itself uses.

Human responsibility

AI-assisted work is reviewed by a person before it goes out. AI output should not be treated as legal, medical, financial, compliance, or other professional advice without review by someone qualified in that field. Using AI as part of a workflow doesn't shift responsibility for the final result away from the people involved. Final decisions, and final accountability, stay with you and whatever licensed or qualified professional is appropriate to the situation, not with a piece of software.

Contact

Questions about how this applies to your specific project are welcome. Reach out at [email protected] and we can talk through what makes sense before any work starts.