WhatsApp users can now check their unread messages with the assistance of Meta AI-powered Message Summaries. They are being rolled out in the United States in English and are created by hitting the unread messages symbol in a chat. By default, these summaries are deactivated.
WhatsApp has unveiled Message Summaries, a new AI-powered tool that summarises unread texts using Meta AI. This feature blocks messages from being viewed by Meta and by others in the chat by using Private Processing technology. Additionally, users can use Advanced Chat Privacy to identify particular texts that require summarisation. The feature, which is by default disabled, is accessible in both individual and group chats. Meta AI creates it in a list view with bullets when you tap the Unread messages icon.
“We’ve all been there – rushing between meetings, catching up after a flight without Wi-Fi, or simply having too many chats to catch up on. Sometimes, you just need to catch up on your messages quickly. That’s why we’re excited to introduce Message Summaries, a new option that uses Meta AI to privately and quickly summarise unread messages in a chat, so you can get an idea of what is happening, before reading the details in your unread messages,” the blog post said.
Also Read: You can now shift the position of the Google Chrome address bar on Android devices
Those who have too many chats to read, need to catch up on meetings fast, or are flying without Wi-Fi, will find this function especially helpful.
In This Article
What is Private Processing?
Meta has created a new technology called Private Processing that allows people to use AI in a way that protects their privacy. Users can instruct AI to handle their requests in a safe cloud environment thanks to the confidential computing architecture, which is based on a Trusted Execution Environment (TEE). This maintains WhatsApp’s fundamental privacy promise, which states that no one can view or share private conversations, while enabling users to take advantage of AI features. Fundamental criteria for private processing include verifiable transparency, enforceable guarantees, and secrecy. While in processing or transit, it must stop other systems—like Meta or WhatsApp—from accessing user data. The system must fail closed or become publicly discoverable through verified transparency if the confidentiality processing assurance is changed.
How Private Processing Works?
A secure cloud environment called private processing enables AI models to process and analyse data without exposing it to unauthorised parties. In order to verify future requests from legitimate WhatsApp clients, it gathers anonymous credentials, retrieves HPKE encryption public keys from a third-party CDN, connects the user’s device to a Meta gateway via an OHTTP connection through a third-party relay, and creates a Remote Attestation + Transport Layer Security (RA-TLS) session between the user’s device and the TEE. To make sure the client connects to code that meets the verified transparency requirement, the attestation verification phase compares measurements to a third-party ledger.
Also Read: Gemini Robotics On-Device model launched, may adapt to new tasks
Once the session is established, the device sends a request to Private Processing, which is completely encrypted using an ephemeral key that WhatsApp and Meta are unable to decode. Without keeping any messages, Private Processing processes data in a confidential virtual machine (CVM) and returns the processed data to the user’s device, encrypted with a key that is only accessible by the device and the Private Processing server of the user’s preference.
Besides this, last month WhatsApp added new functionalities to its Status feature, allowing users to include collages, music, and stickers, similar to Facebook and Instagram’s Stories.
Layout allows users to make collages of up to six photos, whilst Music lets them post a favourite song as a status or turn it into a music sticker. Photo Stickers allow users to make stickers from their images, while Add Yours allows them to invite others to join a conversation, share their responses, and hear what their community has to say.
