Privacy Policy

This policy covers two things that are easy to confuse, so we separate them explicitly.

As a website,EyesInAI publishes benchmark data about AI models. If you are reading the site, we collect very little about you, and the sections “What we collect” and “What we don't do” describe all of it.

As a service operator, Green Package Pro LLCruns assistants and model-routing infrastructure for client businesses. In that role we process conversations that pass through those systems on the client's behalf. If you are talking to an assistant operated by one of our clients, the section “Data we process for client businesses” is the one that applies to you — and the client, not EyesInAI, decides what data goes into their assistant.

• Analytics. We use Google Analytics to understand which pages are useful — aggregate traffic only. We make no attempt to identify you personally.
• Contact form. If you use the contact form, we receive the name, email, and message you send, solely to reply. We don't add you to any list.
• Sign-in, for client accounts only. The client portal uses Google sign-in via Supabase. We receive the email address and account identifier associated with that account, and use them only to authenticate you and authorize what your account may see.

No accounts required for browsing, no tracking pixels beyond analytics, no selling or sharing of data, no advertising networks. The benchmark data on the site is about AI models — not about you.

We do not use client conversation content to train AI models, and we do not sell it or use it for advertising.

When a client business runs an assistant with us, the questions their users ask and the answers given pass through our systems. For a client whose assistant answers over their own business records, that content can include information about theircustomers — people who never interacted with EyesInAI directly. We treat that content as the client's data, held on their behalf.

Why we keep a copy at all. To choose the most accurate and cost-effective model for each request, we record what was asked, what was answered, which model answered, how long it took, what it cost, and how the answer scored. This is what lets us move work to a cheaper model without losing quality, and identify common requests that can be answered without calling an AI model at all.

That conversation content is automatically deleted after 30 days. What remains afterwards is the measurements — which model, what cost, how long, what score — with the words removed. Those measurements contain no personal data. The full rules, table by table, are in the Data Retention Policy.

Who else sees it. To generate an answer we send the request to AI model providers. They process it to return a result and operate their own retention practices, which are governed by their own agreements and are outside our control. Every one of them is named in the Subprocessors list.

All communication with our systems is encrypted in transit using HTTPS. Client conversation records are stored in databases where access is denied by default and reachable only by our own service credentials; no public route exposes them. Access to client data by our staff is limited to what is needed to operate and support the service.

If you contacted us through the site and want that message deleted, email [email protected] and we will remove it.

If you interacted with an assistant operated by one of our clients, the client controls that data. Send your request to them, or to us at [email protected] and we will route it to the right client and assist them in fulfilling it. Requests are handled subject to records we are required to retain for legal, accounting, or security reasons.

The site and the service are not directed to children and we do not knowingly collect information from them.

If this policy changes materially we will update the effective date at the top and, for changes that affect client data handling, notify client account contacts directly.