In an Indian context (as well as for a few other countries including UAE and Mexico), the app and the launch timing gains even more relevance because there is confirmation the Ray-Ban Meta glasses will be launching in India, in the coming weeks. The AI glasses find foundations with Meta AI, including the new Live Translation feature, messaging via Instagram, and contextual conversations. This isn’t the first time Meta AI is available on phones — last summer, Meta had unlocked an AI layer across its popular WhatsApp, Instagram, Facebook and Messenger apps.

The big pitch with the new app seems to be a Discover feed, which places how other users are interacting with Meta AI, in a sort of a timeline. At this time, we can say these are mostly image generations by other users — for now, users have an option to share (or not share) to this feed, for each prompt. The relevance of this may vary for users, but Meta AI becomes the first standalone AI app to add a social media feed as a context to discovering generative AI usage (barring of course the example of xAI’s Grok, which also sits within X).

“Meta AI uses Llama 4 to help you solve problems, navigate your daily questions, and better understand the world around you. With the ability to search across the web, it can help you get recommendations or deep dive on a topic,” the company says, in a statement. For Meta, the Llama 4’s strengths may prove key as it breaks new ground in the search for more users. Some confidence may stem from how well its predecessor model, the LLaMa-3.1-405B passed the Turing Test (a benchmark proposed by Alan Turing in the 1950s to assess whether a machine can exhibit intelligent behaviour indistinguishable from humans) with a score higher than 50%.

There’s convenience to having a separate app, more so with wider capabilities such as full-duplex conversations. Meta needed this. OpenAI’s ChatGPT, Google Gemini, Microsoft Copilot, xAI’s Grok, Anthropic, Perplexity and even DeepSeek, all have had dedicated smartphone apps, for a while. There is diversity in terms of subscription pricing — Meta AI currently does not have a paid tier, which may help with initial traction as curious users begin to use generative AI. Google too have unlocked much of Gemini Live for the free tier subscribers, as has xAI with Grok 3.

In a developing theme over the past few months, Google Gemini is finding deeper integration within a large number of newer Android phones, as well as OpenAI is providing the foundations for Nothing’s AI efforts with the Nothing OS 3.0 such as Essential Space, in phones such as the Phone (3a) series and the new CMF Phone 2 Pro.

Days earlier, Google has released a smarter Gemini Live conversational AI as part of a significant update that also includes video generation using the Veo 2 video model as well as support for more Indian languages. New battlelines are drawn in the AI landscape, almost daily.