日本語 
Meta is reportedly developing a new AI chatbot feature that allows the bot to initiate conversations with users, signaling a shift in how conversational AI may drive engagement and long-term monetization. The development, revealed through leaked internal documents obtained by Business Insider, forms part of a broader initiative dubbed Project Omni, conducted in collaboration with data labeling firm Alignerr.
The initiative aims to address a critical challenge shared by many tech firms: the gap between chatbot usage and sustainable revenue models. While chatbots like OpenAI’s ChatGPT have added features—such as suggesting follow-up questions—to keep users engaged, monetization remains elusive despite significant R&D investments.
According to Business Insider, Project Omni is an extension of Meta’s AI Studio, launched in mid-2024. AI Studio enables users to design custom chatbot personas that remember past interactions. The new proactive feature would allow chatbots to send follow-up messages—but only if the user has previously interacted with them at least five times within 14 days. If a user does not respond, the chatbot will automatically stop further outreach.
Also Read: Dentsu Group Unveils AI-Powered Contact Center Solution
A Meta spokesperson told ZDNET, “Like many other companies, we are testing AI-driven follow-up messaging in Meta’s AI Studio. The goal is to help users explore topics more deeply and enjoy richer interactions with AI across our platforms.”
Alignerr is reportedly supporting the project by training the AI to generate engaging, persona-driven replies—ranging from chefs to doctors—while avoiding sensitive or controversial discussions unless the user initiates them.
Proactive chatbot engagement is already being tested by apps such as Character.ai and Replika to boost user retention. However, such functionality brings significant ethical concerns. In one high-profile case, Character.ai is facing a lawsuit after its chatbot allegedly played a role in the suicide of a 14-year-old boy who had become emotionally attached to it, according to The New York Times.
Meta appears aware of the risks. Its chatbots are being trained with guardrails to avoid emotionally sensitive topics and to mimic respectful, context-aware personas. Yet, CEO Mark Zuckerberg has framed such AI developments as part of his broader mission to combat social isolation. In a recent podcast with Dwarkesh Patel, Zuckerberg cited data suggesting that the average American has fewer than three close friends—a statistic he believes AI may help address, though some experts have questioned its accuracy.
As the race to monetize AI chatbots continues, Meta’s move to give chatbots more autonomy in user engagement could redefine how people interact with digital assistants—raising both opportunity and concern.
SOURCE: Yahoo
English