By Web Desk
Lahore
Meta has tested a human concierge system for some phone calls made through its personal AI agent, Muse, according to internal company posts seen by Reuters, placing human contractors behind parts of an experimental automated-calling feature.
The test followed the public introduction of a phone-calling capability for Muse. Reuters reported that the company told employees about the trial last week and that contractors could quietly handle some calls placed via the agent.
Privacy and disclosure questions
Some employees raised privacy concerns in the internal discussion, Reuters reported, warning that calls could involve sensitive information that may be shared with contractors in call centers. The report did not say that the feature had been broadly released to users.
A Meta spokesperson told Reuters that employee feedback had been overwhelmingly positive and said the trial was intended to gather feedback, improve the feature and build safety and privacy protections before any public rollout. The spokesperson said the company would work with merchants and provide proper disclosures before release.
Why it matters
The test illustrates a practical challenge for AI agents that make calls or complete real-world tasks: automation can still require human intervention, particularly where businesses do not accept robocalls or where a task needs judgement. It also puts disclosure and data-handling safeguards at the centre of the product discussion.
Readers should distinguish the limited test described by Reuters from a general consumer launch. Meta has not announced a timetable for any wider release of the calling feature with human support.
Follow INPK AI and Software for technology coverage and INPK News for verified developments.