ChatGPT can now send texts for you with new Apple Messages plug-in
OpenAI has launched an Apple Messages plug-in for ChatGPT, enabling users to manage, analyze, draft, and send text messages directly through the AI chatbot. This new feature, while offering convenience, also raises privacy concerns regarding user data.
Intelligence analysis by Gemini 2.5 Flash
OpenAI has introduced a new plug-in that integrates ChatGPT with Apple Messages, allowing users to leverage the AI for various messaging tasks like sorting, drafting, and sending texts. The company highlights professional and personal benefits, but the integration also brings forth significant privacy questions, particularly concerning how user message data is handled.
Imagine you have a super smart robot helper named ChatGPT. Now, this robot can connect to your phone's text messages, like a helpful assistant. It can read your texts, help you find old ones, or even suggest what to say back to your friends. But just like you wouldn't let a robot send a text without you checking it first, you need to be careful because it's looking at your private chats!
Analysis
Apple Messages Plug-in
The new Apple Messages plug-in for ChatGPT allows users to connect their iMessage inbox directly with the AI chatbot. This integration offers a range of functionalities, including the ability to sort, analyze, and edit messages. Users can also ask ChatGPT to suggest follow-up messages, delete existing messages, or search through their message history for specific information. The plug-in is designed for both personal and professional use, extending its utility beyond casual conversations to potentially assist with work-related communications.
The commercial advertising the plug-in showcased a user requesting the chatbot to draft follow-up messages based on previous day's interactions. This demonstrates the AI's capability to understand context and generate relevant responses, aiming to enhance user productivity and communication efficiency. The integration with Codex and ChatGPT Work further underscores its potential for professional applications, suggesting a broader scope for AI-assisted communication management.
OpenAI
OpenAI, the developer behind ChatGPT, asserts that the benefits of this plug-in are numerous, aiming to streamline digital conversations for its users. The company has stated that the plug-in runs locally on a user's machine and does not create an index of all messages, attempting to address initial privacy concerns. However, the specifics of what "runs locally" truly entails and the extent of data processing remain somewhat ambiguous.
Furthermore, OpenAI has issued a warning to users regarding the feature that allows ChatGPT to send messages on their behalf. The company explicitly discourages enabling "persistent approval" for message sending, emphasizing that doing so removes the user's final opportunity to review a message before it is dispatched by the AI. This highlights a critical aspect of user control and the potential risks associated with fully automating personal communications.
Privacy Questions
The introduction of a plug-in that grants an AI access to a user's personal message history inevitably raises significant privacy questions. While OpenAI claims the plug-in operates locally and avoids indexing all messages, the precise implications of these technical details are not immediately clear to the average user. The potential for sensitive personal data to be processed, even locally, by an AI model necessitates a deeper understanding of its operational mechanics and data handling policies.
TechCrunch has reportedly reached out to OpenAI for more detailed information regarding these privacy aspects, indicating the ongoing need for transparency. The balance between convenience and data security is a central theme, as users weigh the benefits of AI-powered message management against the risks of exposing their private conversations to an external system. The company's warning about persistent approval also underscores the importance of user vigilance in managing AI permissions, especially when it involves sending communications on their behalf.
Key points
- OpenAI launched an Apple Messages plug-in for ChatGPT.
- The plug-in allows ChatGPT to sort, analyze, edit, draft, and send text messages.
- It is designed for both personal and professional use, integrating with Codex and ChatGPT Work.
- OpenAI claims the plug-in runs locally and doesn't index all messages, but privacy specifics are unclear.
- Users are warned against enabling persistent approval for message sending to maintain review control.
The integration could significantly boost user productivity by automating message sorting, drafting, and searching, making digital communication more efficient for both personal and professional use. It offers a powerful tool for managing overwhelming inboxes and ensuring timely, contextually relevant responses.
The plug-in raises substantial privacy concerns, as the specifics of how user message data is handled locally are not fully transparent, potentially exposing sensitive personal conversations. There's also a risk of users inadvertently sending inappropriate or unintended messages if they enable persistent approval for the AI.
