Meta expands AI assistant to more countries and languages amid intensifying AI battle

Meta announced plans to roll out its AI-powered chatbot, Meta AI, in 21 new markets, intensifying its competition with OpenAI’s ChatGPT and Google’s Gemini.

The company, which now has nearly 500 million monthly active users, will launch Meta AI this week in the UK, Brazil, and several countries across Latin America and Asia.

“It’s coming soon to the Middle East for the first time, too,” the company said in a statement. “We’re adding support today for Tagalog and for more languages in the coming weeks, including Arabic, Indonesian, Thai, and Vietnamese.”

With this phased expansion, Meta expects Meta AI to reach 43 countries and support over a dozen languages, aiming to become the world’s most widely used generative AI assistant by the end of 2024.

The company also announced the rollout of Meta AI on Ray-Ban Meta glasses in the UK and Australia, enabling more smart glasses users to interact with Meta AI hands-free and on the go.

Implications for competition

AI is redefining how businesses and the public interact with technology, as tools like ChatGPT and Microsoft’s Co-pilot become more enterprise-focused, while Google’s Gemini, embedded in Android, extends its reach as a default feature.

“Now, with Meta AI expanding into various regions, I think we are going to see another interesting proposition for public AI,” said Faisal Kawoosa, founder and chief analyst at Techarc. “Especially since some of their apps, like WhatsApp and Instagram, are extremely popular — WhatsApp being the default chat app for billions worldwide.”

The critical factor, however, will be the frequency of user interactions with Meta AI each day. In the initial stages, Meta could benefit from billions of potential queries to further refine its AI, enhancing the platform’s overall appeal.

“While it might be used at scale, whether it will replace Google’s Gemini, which is also on billions of devices, or compete with OpenAI and other assistants remains to be seen,” said Neil Shah, partner and co-founder at Counterpoint Research. “Meta needs to make Meta AI more useful, integrated, and accurate to ensure users don’t leave its platforms for external searches or other tasks.”

Language and expansion challenges

Notably, Meta has not included Europe in its expansion plans, likely due to the region’s stringent regulatory challenges — a hurdle other AI companies have also encountered.

However, even with its current expansion, Meta may encounter challenges, as many languages beyond English are largely untested, according to Pareekh Jain, CEO of Pareekh Consulting.

“Different languages present unique issues, but there’s also a big opportunity to capture new markets,” Jain said. “English is more or less stable now, with a clearer understanding of the issues, but for other languages, we have less data to train on. But Meta already has a strong presence in various languages through platforms like WhatsApp, so they’re well-placed to leverage that.”

This also raises concerns about user data privacy, as the success of Meta AI will hinge on its ability to personalize and localize the model for individual users — a process that necessitates data collection. “If this is the direction Meta is heading, the key concerns for regulators will be how Meta uses user data to train the AI and the implications for user privacy and overall transparency,” Shah added.