Why Meta is testing a ‘human concierge’ for its personal AI agent Muse

Meta is reportedly testing a “human concierge” for its new personal assistant Muse as the company seeks ways to combine AI capability with human support quietly handling certain phone calls through the digital agent.

The update had earlier been shared by Facebook and Instagram owners after the company introduced an outbound calling tool for the public. In light of the current announcement, some employees have raised confidentiality risks concerning manual call processing, warning that it could inadvertently leak sensitive information to contractors.

According to a Meta spokesperson, the response from employees has been exceptionally favourable and they added that the purpose of the test was to gather feedback to optimize safety, privacy and feature performance ahead of the public launch.

Furthermore, the spokesperson Daniel Roberts stated that Meta is closely working with merchants to refine the calling features and will only be launched upon completion and accompanied by all necessary legal disclosures.

Muse Image is Meta’s visual representation of an inaugural image-focused AI model from its Superintelligence Labs which the company confirmed uses advanced reasoning to master intricate concepts into high-quality creations for sharing across its platforms.