Meta AI App Debuts with a Social Edge to Take on ChatGPT

Meta has launched a dedicated stand-alone AI app, marking a major step in its mission to compete directly with industry giants like OpenAI’s ChatGPT, Google’s Gemini, Anthropic’s Claude, and Elon Musk’s Grok. Unveiled at the LlamaCon event in Menlo Park, California, the app is powered by Meta’s advanced Llama 4 model and aims to deliver a highly personalized and socially engaging AI experience.

Unlike its competitors, Meta is capitalizing on its years of social data to enhance user experience.

The company stated the AI can “[draw] on information you’ve already chosen to share on Meta products,” such as your profile, preferences, and online interactions.

This approach enables the AI to tailor its responses based on individual user data, making recommendations or advice more relevant. Users in the U.S. and Canada can also add specific preferences, like dietary restrictions, which the AI will remember in future interactions.

Social Twist to AI

A standout feature of the app is its Discover feed, which allows users to share AI-generated content with friends. This opt-in feature displays prompt-by-prompt interactions and supports likes, comments, remixes, and shares.

Meta believes this feed could fuel creative trends and educate users on AI’s potential in a more interactive, community-driven environment. As Meta’s VP of Product, Connor Hayes, explained, the idea is to show “people what they can do with it.”

Integrated Yet Independent

While Meta AI has already been integrated across Facebook, Instagram, Messenger, and WhatsApp, the new app replaces the existing View app for Meta’s Ray-Ban smart glasses. The revamped app not only supports the glasses but also serves as the primary hub for the AI assistant, offering an intuitive and dedicated interface for daily use.

Voice Mode Innovation

The new app highlights a beta version of voice interaction powered by full-duplex AI, enabling “rich synchronicity in the form of quick and dynamic turn-taking, overlapping speech, and backchanneling.” This makes conversations feel more human-like.

Both standard and full-duplex voice modes are currently available in the U.S., Canada, Australia, and New Zealand.

AI Meets Hardware

Meta’s AI roadmap blends software and hardware. The Ray-Ban smart glasses, for example, already support real-time translation and object recognition. Later this year, Meta plans to launch an upgraded version of these glasses with a built-in heads-up display.

Meta’s CEO Mark Zuckerberg has expressed confidence in the scalability of Meta AI, predicting that 2025 “is going to be the year when a highly intelligent and personalized AI assistant reaches more than 1 billion people.”

Already, Meta AI boasts around 700 million monthly active users, a sharp increase from 600 million in December 2024.

As Meta prepares to release its quarterly earnings and continues its $65 billion investment in AI infrastructure, this new app may prove to be a cornerstone in establishing Meta as a formidable leader in the generative AI space.