Feeling Included by AI: Belonging and Authenticity in Human-AI Interaction

In the rapidly expanding world of artificial intelligence, a significant portion of research has focused on the mechanics of access: who can get online, who has the hardware, and whether the software is affordable. This is the foundation of digital inclusion, a concept that ensures people are not left behind simply because they lack the tools to participate. However, as machines become more conversational and capable of mimicking human social cues, a different kind of question has begun to emerge. It is no longer enough to ask if a person can log in; we must ask how that person feels while they are there. When a voice assistant responds to a query or a chatbot offers advice, does the user feel heard, respected, and free to be themselves? This shift moves the conversation from the structural to the psychological, exploring whether technology can make a human feel like they belong, or if it merely functions as a cold, efficient tool.

A team of researchers set out to investigate this feeling of inclusion within human-AI interactions, focusing on two specific emotional states: belonging and authenticity. Belonging, in this context, is the sense that the system cares about you, remembers you, and treats you as a valued participant. Authenticity is the freedom to express your true thoughts and feelings without fear of judgment or correction. The researchers wanted to know if these feelings mattered. Specifically, they asked whether feeling included by an AI system leads to greater satisfaction with the technology and a stronger sense that the interaction was fair and respectful. To find the answer, they did not rely on a single group of people or a single type of machine. Instead, they gathered data from two distinct populations: 219 users of Amazon Alexa, a voice-activated assistant designed for quick tasks, and 577 users of ChatGPT, a text-based generative AI capable of long, complex conversations.

The study revealed a clear and positive connection between feeling included and how users evaluate the technology. For both groups of people, those who felt a sense of belonging and who felt they could be their authentic selves reported higher levels of satisfaction. They also perceived the interaction as more just, meaning they felt the AI treated them with dignity and fairness. This suggests that the emotional quality of the exchange is just as important as the technical accuracy of the answer. When an AI system responds with empathy, remembers past details, or adopts a non-judgmental tone, it does more than just complete a task; it fosters a relationship that makes the user feel better about the experience.

However, the researchers also discovered that not everyone experiences these interactions in the same way. By analyzing the data patterns, they identified distinct groups of users who react differently to the same technology. In the group of Amazon Alexa users, the data split into two clear profiles. One group, which the researchers called "reserved users," made up about 45 percent of the sample. These individuals reported lower scores for belonging, authenticity, and satisfaction. They seemed to use the device primarily for its utility, treating it as a functional tool without expecting a deep emotional connection. The other group, labeled "engaged users," made up the remaining 55 percent. These users reported high levels of feeling included and satisfied, suggesting that for them, the voice assistant was more than just a command center; it was a source of genuine interaction.

The picture was more complex for the ChatGPT users. Because text-based generative AI allows for richer, more varied conversations, the researchers found six different user profiles, though only three stood out as truly distinct in their experiences. One large group, representing about 33 percent of the users, was described as "utility-oriented." These individuals, like the reserved Alexa users, seemed largely indifferent to feelings of belonging or authenticity. Their satisfaction came from the system's ability to solve problems and perform tasks efficiently, regardless of how the AI made them feel. Another group, the "engaged users," made up about 17 percent of the sample. These individuals felt a strong sense of connection, high authenticity, and deep satisfaction, viewing the AI as a supportive partner. A third group, the "balanced users," constituted the largest segment at roughly 33 percent. These users appreciated the value of the AI and felt a moderate sense of inclusion, but they were also aware of the system's limitations. They enjoyed the interaction but did not expect the same level of emotional depth as the fully engaged group.

These findings challenge the idea that all users want the same thing from their technology. While some people are happy with a machine that simply works, others crave a sense of connection and validation. The study suggests that for AI developers and organizations, the path forward involves more than just improving speed or accuracy. To truly include users, systems must be designed to foster a sense of belonging and allow for authentic self-expression. This means creating interfaces that are empathetic, non-judgmental, and responsive to individual needs. Whether through a voice assistant that remembers a user's preferences or a chatbot that adapts to their emotional state, the goal is to ensure that the interaction feels fair and human. As artificial intelligence becomes woven into the fabric of daily life, from schools to workplaces, understanding these emotional dimensions is essential. It is not enough to build systems that are accessible; we must build systems that make people feel they truly belong.