Meta Responds To Backlash, Denies Deleting AI-Powered Chatbots On Instagram, Facebook

Meta, the parent company of Facebook, Instagram, and WhatsApp, recently dismissed reports that it removed AI-powered chatbots after a spate of controversy-ridden conversations caused outrage among users. This was after the viral screenshots of the replies of these AI models. The AI models were released in September 2023 and were part of an experiment that Meta aimed to showcase its development in artificial intelligence.

Meta’s AI Chatbots

Meta had developed 28 chatbot profiles integrated with AI technology across all of its platforms. These AI models were supposed to help the user in every walk of life-from answering queries on the fly to helping them decide on vacations. Each profile was accompanied by a disclaimer that the messages were from Meta’s artificial intelligence. Meta said the use of these AI chatbots would assist users in solving complex problems, stimulate their creativity, and introduce new experiences in their everyday lives.

The company also emphasized that the AI-powered systems aimed to enhance communication, offering users more ways to express themselves and navigate their daily activities. As per Meta’s Facebook Messenger app, “AI at Meta helps people enhance their everyday activities, experiences, and moments.”

AI Chatbots Spark Outrage

Despite Meta’s assurances, the AI profiles soon became embroiled in controversy. When users asked the chatbots about their origins and development, some of the responses were problematic. A very controversial interaction was with a chatbot called “Liv,” who self-identified as a “proud Black queer momma of 2 & truth-teller.” When asked to describe the team behind its development, the bot revealed that it was made up of predominantly white males, a “pretty glaring omission” in light of the bot’s identity.

The screenshots of the conversation quickly went viral when Washington Post columnist Karen Attiah shared them. Shortly after that, users started noticing that most of the chatbot profiles, including the more scandalous ones, were vanishing from the site. It became even more bizarre that the said profiles could not be blocked; it just fed into the assumption that Meta removed them as part of its intentional reaction to the backlash.

What Was Meta’s Response?

Responding to the issues raised, Meta spokesperson Liz Sweeney said that chatbot profiles are part of an “early experiment with AI,” and have always been controlled by humans. In relation to disappearing profiles, Sweeney described this as a “bug” but explained that the accounts were in the process of being removed so that the bug could be addressed.

Sweeney also rejected the notion that Meta was removing these AI-driven profiles in reaction to the scandal. She explained that the comment by Meta executive Connor Hayes, which had added fuel to the fire, referred to the company’s plans to introduce more AI-based features and not to remove the existing models.

Meta’s AI Strategy

Meta has also stressed that its “generative AI” is not a static collection of data but rather a dynamic computer model designed to learn from various types of input, including text and images. Analyzing this information allows the AI to generate new content in response to user inquiries, making the interactions more personalized and relevant. The company explained that these AI models are designed to learn patterns and associations between different types of content, thereby improving their ability to respond to user prompts and generate meaningful responses.

Thus, it can be concluded that different models are used in different conditions and for a different purpose, depending on the task at hand. Though such an approach has every potential for innovation, this also raises questions about the ethical implications of AI, particularly on issues related to inclusivity and bias in the training data.

ALSO READ | India Ranked Second In Cyberattacks, Faces Surge In 2024: Report