Ghost in the Machine: Building an Open-Source Privacy Boundary for AI Agents

Description and use

<p>AI agents are useful precisely where the stakes are high: customer support, financial operations, clinical workflows, and internal tooling.</p> <p>That is also where they can accidentally send the wrong thing to an external model.</p> <p>A bank-transfer agent may receive a customer’s name, phone number, account number, BVN, or NIN. A support agent may handle email addresses and sensitive tool results. Most AI integrations still treat the model request as the boundary: collect data, serialize it, send it upstream.</p> <p>I wanted to invert that assumption.</p> <p><strong>Aegis is an open-sou

Problem

<p>AI agents are useful precisely where the stakes are high: customer support, financial operations, clinical workflows, and internal tooling.</p> <p>That is also where they can accidentally send the wrong thing to an external model.</p> <p>A bank-transfer agent may receive a customer’s name, phone number, account number, BVN, or NIN. A support agent may handle email addresses and sensitive tool results. Most AI integrations still treat the model request as the boundary: collect data, serialize it, send it upstream.</p> <p>I wanted to invert that assumption.</p> <p><strong>Aegis is an open-sou

Outcome

<p>AI agents are useful precisely where the stakes are high: customer support, financial operations, clinical workflows, and internal tooling.</p> <p>That is also where they can accidentally send the wrong thing to an external model.</p> <p>A bank-transfer agent may receive a customer’s name, phone number, account number, BVN, or NIN. A support agent may handle email addresses and sensitive tool results. Most AI integrations still treat the model request as the boundary: collect data, serialize it, send it upstream.</p> <p>I wanted to invert that assumption.</p> <p><strong>Aegis is an open-sou

Evidence

Ghost in the Machine: Building an Open-Source Privacy Boundary for AI Agents

Source