Is data encrypted before leaving your machine?

Yes. The Privatemode proxy runs locally and encrypts all data before it leaves your machine. It also performs remote attestation to verify the integrity of the Privatemode service, so every request is protected end to end by confidential computing.

Does Privatemode store or retain workflow data?

No. Privatemode never learns from your data. All requests are processed using confidential computing. Prompts and responses are not saved once a request completes; only operational metadata such as token usage, timestamps, and request status is kept for billing and monitoring.

Which n8n AI nodes are supported?

All nodes that use the OpenAI Chat Model sub-node work with Privatemode, including AI Agent, Basic LLM Chain, Sentiment Analysis, Summarization Chain, Text Classifier, Question and Answer Chain, and Information Extractor. The Embeddings OpenAI sub-node works with the same credential and the model `qwen3-embedding-4b`, so vector stores stay on the proxy too.

How does confidential computing affect workflow performance?

Privatemode uses confidential computing hardware to minimize overhead. Confidential computing adds a small overhead. Response times are comparable to conventional AI services, and your data stays encrypted in memory during processing.

Confidential AI workflows with Privatemode and n8n

n8n workflows call models on schedules and webhooks, thousands of times. Pointing its AI nodes at Privatemode keeps that traffic encrypted before it leaves and encrypted in memory while it is processed.

Introduction

An n8n AI Agent workflow whose chat model node is connected to Privatemode

The challenge

Every system n8n connects becomes AI input

n8n workflows pull data from your CRM, mailbox, ticketing system, databases, and document stores, then hand it to an AI node to classify, summarize, or extract. Customer records, contracts, and internal documents travel with it. With a conventional AI provider, all of that is processed in plaintext on the provider's servers, and it happens on every run, without anyone looking at what was sent.

The solution

The same workflows, a provider that cannot read them

Point n8n's AI nodes at the Privatemode proxy running next to it: one credential covers the AI Agent, the chains, and the embeddings behind your vector stores, so the switch is made once rather than per workflow. Every request is encrypted before it leaves and processed in a hardware-isolated environment whose memory stays encrypted, so by design neither the infrastructure provider nor Privatemode can read your data.

Benefits

Why use Privatemode AI with n8n?

Works with every AI node

End-to-end confidentiality

Learn how

Free tier available

How to get started

Set up confidential agents in n8n in just a few steps

Get your API key

If you don't have a Privatemode API key yet, you can generate one for free here.

Run the Privatemode proxy

The proxy verifies the integrity of the Privatemode service using confidential computing-based remote attestation. The proxy also encrypts all data before sending and decrypts data it receives.

Set up an AI agent in n8n

Follow the official n8n AI Agent documentation to set up an AI agent. Since Privatemode adheres to the OpenAI interface specification, you can use the standard OpenAI Chat Model node and configure its credentials to connect to Privatemode.

In addition to the AI Agent node, the OpenAI Chat Model can be used in several other powerful AI nodes, including:

For vector stores, use the Embeddings OpenAI sub-node with the same credential and the model qwen3-embedding-4b, so embeddings are computed through the proxy as well.

Adjust credentials

To ensure that all data is sent through the Privatemode proxy, set the following credentials for your OpenAI Chat Model node:

API Key: Set this to a placeholder value, such as "placeholder". Authentication is handled by the proxy.
Base URL: Set this to the /v1 route of your Privatemode proxy, for example, "http://localhost:8080/v1".

Set the parameters of the node

Set the parameters of the OpenAI Chat Model node to use the correct credentials and select the model you want to use.

Done!

You’ve successfully integrated Confidential AI into your workflow.

FAQ

Frequently asked questions about using Privatemode with n8n

No structural changes are needed. You reconfigure the OpenAI Chat Model node credentials to point to the Privatemode proxy by changing the Base URL field. All AI nodes that use this model, including AI Agent, Basic LLM Chain, and Summarization Chain, are protected by confidential computing automatically.