Is Meta Relying Too Much on Artificial Intelligence?

TECH NEWS – It’s not just because Mark Zuckerberg’s company is suspending or locking out innocent users en masse for overreliance on artificial intelligence; it’s also because Zuck’s company wants to rely on AI chatbots…

 

Tech giants are constantly focusing on using AI technology to improve processes for users and transform the way communications are conducted. Meta is an example of this ambitious approach to technology; it constantly looks to improve the user experience. Now, the company led by Mark Zuckerberg is working on making its chatbot an assistant or even a companion, not just an AI tool. Meta is reportedly training its chatbot to initiate conversations based on previous ones and reach out proactively.

Meta is looking for ways to increase user engagement and loyalty, especially as competition increases and companies innovate, offering powerful AI experiences. (We would like to note here that this would require human moderation. It would not just use AI for moderation and would not allow banned or suspended users to be overridden by AI in a matter of minutes. Users shouldn’t have to subscribe to receive customer service that seems less and less human.) Meta understands that conversations often lag, especially in today’s world, where users are busy and communication is difficult to initiate and maintain.

Meta is reportedly collaborating with Alignerr to develop customizable chatbots that can proactively reach out to users, initiate conversations, and track previous interactions. Created using Meta AI Studio, the bots will be able to retain user information, allowing them to engage in more human-like, personalized, and context-aware conversations.

According to the project guidelines, the training project is known internally as Project Omni, and the initiative aims to improve user retention and engagement. This move aligns with Zuckerberg’s efforts to combat loneliness by helping users connect and improve responsiveness. Meta also aims to prevent AI chatbots from becoming a nuisance.

To this end, chatbots will be designed to send follow-up messages within 14 days of initiating a conversation. The company stresses that the bot will not continue sending messages if the user does not respond to the initial follow-up message. Users can choose whether to keep the bots private or share them publicly on their stories or profile.

While this technology is not new—other AI startups, such as Character.AI, offer similar options—this move clearly shows that the tech giant is working to create a more personalized experience for users and is looking for ways to increase engagement.

Source: WCCFTech, Business Insider

Spread the love
Avatar photo
theGeek is here since 2019.

No comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.