# WhatsApp Developing On-Device Call Screening for Android

New feature enables automated transcription of unknown callers to enhance user privacy and control.

- **Source:** [Technology | The Indian Express](https://indianexpress.com)
- **Published:** 2026-10-20
- **Topic:** [Software & Engineering](https://brifing.si/c/engineering)
- **Original article:** <https://indianexpress.com/article/technology/whatsapp-could-soon-let-you-screen-unknown-callers-before-answering-10929401/>
- **On Brifing:** https://brifing.si/s/d86cdbc9-33fc-4d4c-98e1-4ead81cbe4db

## New Call Filtering Architecture

WhatsApp is developing a sophisticated call filtering mechanism for Android, identified in beta version 2.26.41.4. This feature aims to provide granular control over incoming communications from unsaved numbers. By integrating directly into the privacy settings, the platform intends to evolve beyond the current basic silence functionality into a more interactive screening tool.

## Three-Tiered Privacy Management

The proposed system introduces three distinct settings for users to manage incoming traffic. The 'Never' option disables all filtering, while the second setting silences unknown callers without triggering a ringtone. The third, most advanced option, automates the interaction by prompting unknown callers to state their purpose before the user decides to answer.

## Automated Transcription Mechanism

When an unknown number calls, the application will automatically capture the caller's response and generate a text transcript. This allows the recipient to review the caller's intent without engaging in a live conversation. This mechanism effectively serves as a digital gatekeeper, filtering out spam while ensuring legitimate callers can still reach the user.

## On-Device Processing Privacy

To maintain end-to-end encryption standards, WhatsApp plans to perform all transcription tasks locally on the user's device. By utilizing specific language packs, the app avoids sending audio data to external servers. This architectural choice prioritizes user privacy, ensuring that sensitive voice data remains isolated from the company's central cloud infrastructure during processing.

## User-Defined Language Support

The implementation will likely include a configuration menu allowing users to select their preferred language for transcription. This ensures that the automated screening process remains accurate and accessible across diverse linguistic regions. By localizing the transcription engine, WhatsApp aims to provide a consistent experience for its global Android user base during the screening.

## Comparative Industry Context

This development mirrors similar call screening capabilities previously introduced by Apple in iOS 26. By adopting a comparable solution for the Android ecosystem, WhatsApp is aligning its feature set with broader mobile industry standards. This move highlights a growing trend among messaging platforms to integrate native OS-level communication management tools directly into apps.

## Development and Beta Status

Currently, the feature remains in the early stages of development and is not yet accessible to the public or beta testers. WABetaInfo reports that the functionality is still being refined internally. No official release date has been announced, suggesting that further testing is required before a wider rollout to the stable channel.

## Strategic Impact on UX

If successfully deployed, this feature will significantly reduce the friction caused by unsolicited calls. By combining spam prevention with automated screening, WhatsApp is enhancing the overall utility of its voice services. This evolution reflects a shift toward more proactive communication management, empowering users to maintain focus while remaining reachable for important contacts.

## Sources covering this story

- https://indianexpress.com/article/technology/whatsapp-could-soon-let-you-screen-unknown-callers-before-answering-10929401/ — 

---

_Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._
