Meta wants people to use ‘AI agents’ to do daily tasks for them. Here’s why that’s a problem
Meta can’t beat its rivals on raw AI capability. With its new Muse agent, it doesn’t need to
Meta aims to empower users with AI agents that can handle daily tasks like booking flights or shopping. The company's new Muse agent gained popularity after Mark Zuckerberg demonstrated it at a conference. Muse, billed as having "no learning curve," can perform a range of actions, from ordering groceries to booking appointments.
Agents, powered by chatbot-like large language models and reinforcement learning, have evolved from hobbyist projects to personal assistant apps. Meta's entry into the AI agent market marks a shift towards mainstream adoption. Despite spending billions on AI and capital expenditure, Meta's AI models still rank behind those of competitors like Anthropic and OpenAI.
However, Meta plans to offer paid AI services to cater to businesses and professionals. Muse's appeal lies in its user-friendliness, with a built-in privacy feature that runs on a dedicated virtual machine. The agent can be personalized with an anthropomorphic avatar and video chat capability. Despite its promising features, Muse is currently only available in the US and limited to non-sensitive tasks.
As users become more reliant on AI agents for managing online interactions, concerns about privacy and data control arise. Meta's history of data breaches and privacy scandals adds skepticism to the platform's promises. The lack of regulations governing AI agents further amplifies the risk. As Muse gains traction, the question remains: will it become a primary interface for users, posing a threat to Meta's existing position while potentially gathering more user data than social media platforms?
Written by urgent.news from The Conversation AU's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
