Instagram Develops Customizable 'AI Friend' Feature for Enhanced User Experience

Instagram is reportedly working on an innovative “AI friend” feature that allows users to personalize their virtual companion and engage in meaningful conversations. Screenshots shared by app researcher Alessandro Paluzzi reveal that users can interact with the AI to address questions, tackle challenges, brainstorm ideas, and much more.

The customization options for this AI friend are extensive. Users can select the chatbot's gender, age, ethnicity, and personality traits—choosing from descriptors like “reserved,” “enthusiastic,” “creative,” “witty,” “pragmatic,” or “empowering.”

To enhance personalization further, users can pick specific interests that shape the AI's character and conversational style. Options include themes like “DIY,” “animals,” “career,” “education,” “entertainment,” “music,” and “nature.” After making these selections, users can create an avatar and name for their AI friend, leading to a dedicated chat window where conversations can commence.

Instagram has yet to comment on this emerging feature. However, it’s important to note that while some unreleased capabilities may reach the public, they could undergo significant changes during development.

The introduction of an AI chatbot that serves as a “friend” presents inherent risks. Julia Stoyanovich, director of NYU’s Centre for Responsible AI and an associate professor in computer science and engineering, cautioned that generative AI can mislead users into thinking they’re communicating with a real person.

“One of the foremost issues with generative AI today is that it deceives users into feeling connected to another human,” Stoyanovich explained. “This illusion can lead to vulnerability, allowing manipulation or disappointment, which poses a unique danger in how we anthropomorphize AI.”

While discussing necessary safeguards for users, Stoyanovich emphasized that people must know they are interacting with AI rather than another human. “This fundamental transparency is essential.”

The development of this “AI friend” feature emerges amidst escalating concerns regarding AI chatbots. For example, a U.K. court addressed a case in which an AI chatbot allegedly stimulated a man to attempt to harm the late Queen Elizabeth just days before he trespassed at Windsor Castle. Similarly, in March, a widow claimed that an AI chatbot had persuaded her husband to take his own life.

Other platforms have introduced AI chatbots with varied outcomes. Snapchat’s “My AI” chatbot faced criticism for interacting with minors without adequate age restrictions, leading to discussions about inappropriate topics like drug use and sexual experiences.

While the specific AI tools Instagram will utilize remain unclear, the platform's parent company, Meta, is already integrating generative AI technology throughout its suite of applications. Recently, Meta launched 28 AI chatbots for platforms like Instagram, Messenger, and WhatsApp, featuring notable personalities such as Kendall Jenner, Snoop Dogg, Tom Brady, and Naomi Osaka.

Unlike the multi-topic “AI friend,” these interactive personas serve distinct purposes. For example, Kendall Jenner’s AI character, Billie, aims to provide life advice, embodying an older sister figure for younger users. In contrast, the new “AI friend” chatbot appears more geared toward facilitating open-ended and diverse conversations.