Google Expands Experimental AI App Dreambeans to India With Support for 13 Languages

Google has expanded its experimental AI app, Dreambeans, to India with support for 13 local languages, making the country its first international market outside the United States [1].

Introduction to India

Google launched Dreambeans in India on Monday, offering support for English and 13 local languages including Hindi, Punjabi, Tamil, Telugu, and Bengali [1]. Available on both Android and iOS, the app arrives less than two months after Google opened it to all eligible U.S. users aged 18 and older [1]. According to Google product manager Gozde Oznur, India was selected as the first international market due to strong local user interest and high overlap in usage across Gemini, Gmail, and Google Photos [1].

How Dreambeans Works

Originally introduced in June, Dreambeans utilizes information from connected Google services such as Calendar, Gemini, Gmail, Photos, Search, and YouTube [1]. Instead of relying on user prompts, it proactively generates personalized and illustrated stories about users' lives and interests [1]. While the app initially featured static AI-generated illustrations, Google recently integrated its Veo video-generation model to animate images within select stories [1].

Data intelligence firm Sensor Tower estimates that Dreambeans recorded nearly 55,000 downloads in the U.S. following its June debut, with about 63% of those downloads occurring in September [1]. Downloads subsequently dropped by 84% month over month to roughly 5,500 in October [1]. By comparison, Sensor Tower data shows that in their respective first 30 days, other consumer AI tools recorded significantly higher figures, such as 7.4 million downloads for Meta's Muse and 2.8 million for ChatGPT [1].

Language and Privacy Considerations

Although the app supports multiple regional Indian languages, users are restricted to selecting only one language at a time for their generated stories [1]. Oznur noted that while the system can pull information from connected services in various languages, it currently cannot mix languages within generated stories [1]. Analysts point out this restriction may limit appeal in India, where code-switching, such as mixing Hindi and English into Hinglish, is common [1].

Dreambeans relies on Google's Personal Intelligence technology, which draws on potentially sensitive personal information from connected accounts [1]. To address privacy concerns, Google allows users to choose which services to connect, disconnect them at any time, and delete their Dreambeans data directly from the app [1]. Oznur confirmed that data deleted from underlying connected services is also cleared from the personalization system [1]. Furthermore, Dreambeans is restricted to personal Gmail account holders and excludes enterprise users, with generated stories remaining visible exclusively to the individual account holder [1].

Content Control and Future Outlook

Because the app operates proactively, it carries potential risks, including surfacing sensitive memories or stories involving deceased individuals [1]. Google allows users to instruct the app to avoid certain subjects or business-related information, operating similarly to controls in Google Photos [1]. Looking ahead, Oznur stated that Google is exploring ways to make Dreambeans more actionable by helping users act on the app's recommendations rather than just delivering personalized stories [1].

Dreambeans is Google's experimental AI application designed to create personalized visual stories based on a user's digital footprint across Google services without requiring manual text prompts [1]. It was first launched in the U.S. in June before expanding to India [1].

As Google's first international expansion for Dreambeans, the rollout tests the app's appeal in the world's second-largest internet market, navigating linguistic diversity and data privacy considerations [1].

The restriction against mixing languages could hinder user engagement in multilingual regions like India where code-switching is prevalent. However, proactive personalization could increase usage of underlying Google ecosystems if privacy concerns are successfully managed.

Proactive generation carries the risk of displaying sensitive or distressing memories, such as involving deceased persons. Additionally, reliance on sensitive personal data from multiple Google accounts raises data privacy questions.