Confidential by default
Start with a high-level description. What you share stays confidential. What we create for you belongs to you.
Trust & boundaries: good work begins with trust.
You should know how your information is handled, what supports a recommendation, and who is responsible for the work. These expectations belong in the relationship from the beginning.
What you can expect from us.
Clear practices help us work together with confidence.
The right tool for the work
We are model agnostic and work across providers, including OpenAI, Anthropic, Google, and open-source models. deepmoat.ai is an OpenAI Select Partner and a registered partner in Anthropic’s Claude Partner Network; those relationships do not decide the recommendation. Relevant referral or resale interests are disclosed before a commitment.
Reasons you can examine
We connect material claims to sources and explain where evidence ends and judgment begins. A demonstration or announcement is not treated as proof of results in your setting.
People accountable for the work
AI can support research, comparison, and testing. People review the work and own the recommendation. Human judgment remains part of the service.
Respect for client privacy
We do not need to display your name or work to explain our approach. Client information, logos, and results are shared publicly only with approval.
An agreement before the work expands
The scope, fee, access, deliverables, and responsibilities are settled before deeper work begins. Additional work is discussed as a separate commitment.
Respect for professional responsibility
Our work supports your understanding and choices. Your people make every decision; we prepare the ground.
AI helps with the work. People stand behind it.
We use AI where it helps us research, organize, compare, or test. Its output still needs examination: sources can be weak, details can be wrong, and a confident answer can miss the point.
Client material is handled according to the agreed scope and access arrangements. We discuss which information and tools are appropriate before connecting private work to them.
You should be able to ask why.
A useful recommendation comes with an explanation. Ask what we considered, what remains uncertain, and what would cause us to change our view.
You retain the final decision and your organization’s responsibilities. We bring our experience and make our reasoning available, so you can decide with a clearer understanding.
Questions belong at the start.
Have a question about how we would handle your work or information? Raise it at the beginning. It belongs in the conversation.