After Users Abandon Threads, Zuckerberg Introduces Meta’s New Project To Retain Users

So, what is Mark Zuckerberg cooking in Meta’s kitchen?

The newest test in Meta's quest to keep consumers engaged: chatbots with distinct personalities!

In an effort to increase user activity on their platforms such as Threads and Instagram, Meta, the company that owns the social media platforms, plans to release a variety of AI-powered chatbots with distinct personalities as early as next month.

According to three people familiar with the company's ambitions, the digital giant helmed by CEO Mark Zuckerberg has been creating prototypes for chatbots that can have humanlike dialogues with its almost 4 billion users.

Employees have given some of the chatbots "personas" and according to those familiar with them, those "personas" can take on a variety of personalities. According to the source, the corporation has considered releasing one that sounds like Abraham Lincoln and another that gives travel advice in the voice of a cool surfer dude.

According to the company, the rollout of the chatbots is planned for September.

Competing companies have already released chatbots with distinct identities. The Andreessen Horowitz-backed, $1 billion-valued startup Character.ai employs massive language models to simulate the speech patterns of recognizable figures like Tesla CEO Elon Musk and Nintendo character Mario.

This comes as firms compete in the social media space and capitalize on the buzz surrounding artificial intelligence (AI) in Silicon Valley, ever since Microsoft-backed startup OpenAI released ChatGPT in November.
 
So, what sort of AI offerings are Meta and Zuckerberg planning next?
 

What Will Be Meta’s Next Step?

Zuckerberg told analysts that the company would share its AI product road map at its Connect developer event in September.

According to Zuckerberg, AI "agents that act as assistants, coaches, or that can help you interact with businesses and creators" are in the works, and he added, "We don't think that there's going to be one single AI that people interact with."

He has also indicated that the firm is developing an AI-powered productivity aide for employees, as well as an AI-powered customer care representative.
 

Will This Be Another Advance In Generative Artificial Intelligence?

Meta has funded research and development into generative AI, which can generate content such as text, graphics and code. To fuel its chatbots, the company has published Llama 2, a commercial version of a large language model, this month. According to two sources with knowledge of the situation, Meta has been trying to acquire tens of thousands of GPUs as part of constructing the infrastructure to support the AI products. GPUs are chips that are crucial for powering massive language models.

The corporation has dabbled with chatbots on a lesser scale before, and those experiments revealed the aforementioned dangers. Researchers discovered that after the introduction of BlenderBot 2 in 2021, a prior Meta AI model had already begun spreading disinformation. Meta said that it had made the 2022 version of BlenderBot more resilient to such content but users still reported that it provided erroneous information and hate speech.

An insider at Meta has said that the company is planning to implement a system to moderate user questions. The individual continued, "The company may also automate checks on the output from its chatbots to ensure what it says is accurate, and avoids hate speech or rule-breaking speech, for example."
 
Will Meta and Zuckerberg hit the AI jackpot and retain users with their latest gamble on chatbots? Let us know your thoughts in the comments section below!