ChatGPT iMessage Integration for Agencies: Client Communication Workflows

Published August 21, 2026By ABD Legacy LLC
AI tools ChatGPT iMessage integration for agencies AI client communication workflow

On August 20, 2026, OpenAI shipped an Apple Messages plugin for the ChatGPT desktop app on macOS. With permission granted, ChatGPT can read, search, summarize, draft, and send iMessage, SMS, and RCS messages through the Messages app (OpenAI release notes, Codex plugin docs). For agencies running on ChatGPT Work and Codex, this is a new tool in the agentic surface — and a new workflow layer for client communication.

What the ChatGPT Apple Messages plugin actually does

The plugin reads and searches iMessage, SMS, and RCS conversations on the Mac, and can find, summarize, draft, or send messages through the Messages app (9to5Mac, OpenAI). It is available on all plans in the ChatGPT desktop app for macOS, with the functional surface in ChatGPT Work and Codex (MacRumors). Two scope limits matter before you build on it: it does not let you interact with ChatGPT remotely through Messages, and it does not work in regular ChatGPT chats, ChatGPT on the web or mobile, Codex CLI, or the IDE extension (Unite.AI).

Client-communication workflows agencies can build today

These are the concrete, agency-specific workflows the plugin enables. Each one keeps the human in the loop — which is the selling point for agencies that cannot auto-send on behalf of clients.

The approval gate is the workflow design. By default, ChatGPT sends messages only after the user approves the message and its recipients (OpenAI). That makes the practical pattern "AI drafts, human approves" — the correct posture for client-facing communication.

Setup and permissions (what to expect on a Mac)

The feature is opt-in, and setup is a macOS permission flow, not a toggle:

  1. Apple silicon required. The plugin is included only in the Apple Silicon (arm64) build of the ChatGPT desktop app; it does not run on Intel Macs (OpenAI, 9to5Mac).
  2. Grant macOS permissions during the setup permission screen (Engadget).
  3. Full Disk Access in System Settings (Mac privacy preferences) (MacRumors, TechCrunch).
  4. Contact names and automation tools — access to the names of contacts and to automation tools (AppleScript and Accessibility settings) (Engadget, MacRumors).

OpenAI says the plugin runs locally on the user's device (TechCrunch). That is a statement about the plugin runtime — not a statement about what leaves the machine when you ask it to summarize an archive (see the compliance section below).

Limitations agencies should plan around

Client data, privacy, and compliance caveats

The documented control story is real: an OpenAI spokesperson said the plugin does not create an index of a user's texts, ChatGPT reads messages only when the user explicitly asks, and the plugin will not send until the user approves the drafted message (Engadget, TechCrunch). OpenAI told Engadget: "We care deeply about user privacy, and the plugin is designed to be transparent and keep users in control of their Messages data."

The honest caveats for agency compliance: OpenAI has not published exactly which message content leaves the machine when you ask it to summarize or analyze an archive, and group-chat participants are not asked for consent when ChatGPT analyzes a conversation (TNW). There is also no published enterprise/admin data guidance for Work-managed Macs mixing personal and professional archives (TNW). Practical compliance framing for agencies handling client data:

For the deeper security vetting frame, see our AI agency security vetting guide, and for the sibling agent-messaging story, read how Claude Code sessions can now message each other. The ChatGPT-for-agencies thread lives in our ChatGPT Ads coverage, and Work/Codex cost context is in our AI coding agent pricing guide.

Build client-communication workflows that keep a human on the send button.

Browse AI Agencies →

Frequently asked questions

Can ChatGPT read my texts on an Intel Mac?

No. The Apple Messages plugin is included only in the Apple Silicon (arm64) build of the ChatGPT desktop app. It does not work on Intel Macs.

Does ChatGPT auto-send messages?

No. Sending is gated by default: ChatGPT sends a message only after you approve the message and its recipients. You can choose "Allow once" per send, or persistent approval per conversation — which OpenAI discourages because it removes your final chance to review a message before it is sent as you.

Does OpenAI keep an index of my messages?

Per an OpenAI spokesperson, the plugin does not create an index of a user's texts, and ChatGPT reads messages only when the user explicitly prompts it to draw on past messages.

Can my agency turn this off for employees?

Yes. In managed workspaces, administrators can disable Apple Messages through the existing Computer Use control, the same mechanism that governs other desktop-agent capabilities.

Is this available in regular ChatGPT?

No. The plugin works in Codex and ChatGPT Work, not regular ChatGPT chats, and it is not available in ChatGPT on the web or mobile, in Codex CLI, or in the IDE extension.

Sources

Accuracy note: announced Aug 20, 2026 in OpenAI's release notes; "available on all plans" with the functional surface in ChatGPT Work and Codex (per OpenAI's own docs — Engadget phrased availability as Work/Codex users); plugin included only in the Apple Silicon (arm64) build. OpenAI has not published exactly which message content is transmitted for processing, whether Apple was involved, or enterprise admin data guidance for managed Macs — those are open questions, not confirmed facts. Apple was not part of the announcement; Apple sued OpenAI in July 2026 (trade secrets) and previously cut off third-party iMessage access (Beeper Mini, 2024) — background context only, no speculation on outcomes.

Related reading