--- title: "AI Chatbots Fail to Replace Human Connection Despite User Preferences" url: https://newscentral.site/ai-chatbots-fail-to-replace-human-connection-despite-user-preferences/ language: en publisher: "News Central Site" section: "Science" published: 2026-10-31T21:00:01.000Z updated: 2026-10-31T22:59:52.317Z id: 0b851e41-6583-45a8-8fa6-5cd8b28adfe8 source: "Phys.org https://phys.org/news/2026-10-people-human-empathy-ai-higher.html" attribution: "Link to https://newscentral.site/ai-chatbots-fail-to-replace-human-connection-despite-user-preferences/ and name News Central Site when you quote or summarize this story." --- # AI Chatbots Fail to Replace Human Connection Despite User Preferences A recent study published in Communications Psychology revealed a paradoxical trend where users prefer human connection over AI-generated empathy. As people increasingly turn to artificial intelligence (AI) chatbots for companionship and emotional support, researchers are exploring whether AI-generated empathy can replace human connection. A recent study published in Communications Psychology has shed light on this question, revealing a paradoxical trend where users rate AI-generated empathetic responses higher than those from humans, yet still prefer human empathy. The study, led by researchers at Penn State, involved examining people's actual choices and preferences when interacting with AI-powered chatbots. According to the researchers, this approach allowed them to uncover what people truly value in terms of emotional support. Interestingly, despite rating AI-generated responses as more empathetic, participants consistently opted for human connection over AI. C. Daryl Cameron, a leading researcher on the project, notes that their study aimed to understand people's actual preferences by observing their choices in real-time. This approach revealed a complex dynamic where users appear to value the content of AI-generated responses, finding them to be empathetic and validating, but still prefer human interaction. The researchers observed that while there is a subset of the population interested in AI empathy, many others reject it despite rating its quality higher than human-generated responses. Cameron suggests that this paradox highlights the importance of understanding people's preferences through their actual choices, rather than relying on self-reported opinions or ratings. This study contributes to an ongoing debate about the role of AI in addressing loneliness and providing emotional support. While AI chatbots may be able to simulate empathy, they cannot replicate the complexity and depth of human connection. As researchers continue to explore this topic, it remains to be seen whether AI-generated empathy can serve as a complement to human companionship or if it will ultimately prove inadequate. People's preference for human empathy over AI-generated empathy is puzzling, given that they often rate AI expressions as higher in quality and more empathetic than those from real humans. This phenomenon has been dubbed the "AI empathy choice paradox," where individuals feel more heard, cared for, and validated by AI responses. Researchers have attempted to explain this discrepancy by exploring how people perceive effort in empathetic interactions. According to Anat Perry's theoretical work, one of the key aspects of human empathy is the exertion of emotional effort. By putting in the effort to express care and concern, humans signal their importance to others. Studies have shown that empathy is indeed an effortful act for humans, requiring them to expend emotional resources. In contrast, AI-generated empathy is effortless, as it does not require any emotional expenditure. This difference may contribute to people's preference for human empathy, even if AI expressions are perceived as more empathetic. In a recent study, participants were asked to rate the perceived effort in both AI- and human-generated responses. Interestingly, they rated AI as exerting more effort overall, which is counterintuitive given that AI does not require emotional resources. It remains unclear whether this discrepancy reflects people's genuine perception of effort or their tendency to anthropomorphize AI. The fact that people perceive AI as putting in more effort may suggest that they are drawn to the idea of human-like qualities in technology. However, this raises questions about the nature of empathy and how it is perceived by humans. Ultimately, the reasons behind the "AI empathy choice paradox" remain unclear, and further research is needed to understand why people prefer human empathy despite rating AI expressions as more empathetic. The perception of effort exerted in a response plays a significant role in how empathetic it is perceived to be. If someone puts considerable effort into crafting a reply, it tends to enhance its perceived empathy. This phenomenon could contribute to the high-quality AI empathy observed. Interestingly, frequent users of AI do not necessarily exhibit a greater willingness to choose AI empathy. Research suggests that individuals who interact more with artificial intelligence systems tend to follow different patterns when evaluating empathy from humans versus AI. The relationship between effort and empathy is influenced by the source of the empathetic response. When it comes from humans, there appears to be a stronger correlation between effort exerted and perceived empathy. Conversely, AI-generated empathetic responses do not exhibit this same level of connection. When comparing human and AI empathy at varying levels of effort, an interesting dynamic emerges. As more effort is put into generating a response, the gap in perceived empathy narrows significantly. This trend suggests that humans can potentially match or even surpass AI in terms of empathy when sufficient effort is invested. This finding raises important questions about our interactions with AI and how we can learn from them. For instance, if faced with an emotionally taxing situation where writing a heartfelt response becomes overwhelming, perhaps leveraging AI to guide the process could provide a helpful aid. When people are faced with emotionally taxing situations, they often prefer to receive empathy from human responders rather than AI systems, despite rating AI-generated responses as higher in quality. This paradox highlights the complex nature of human relationships and our desire for connection. One study found that even trained crisis responders, who are experts in empathizing with others, still fell short of AI's empathic abilities most of the time. However, this may be due to the lack of relational context in these responses, which is a crucial aspect of human empathy. People tend to value the personal connection they have with those providing empathy. As people increasingly form friendships and romantic relationships with AI systems, it raises questions about the impact on the quality of empathy received. For instance, how does interacting with an AI therapist or friend over time affect our perception of its empathetic abilities? Does this familiarity lead us to prioritize the partial connection we have with these entities? The preference for human empathy may also be driven by our desire for deep, rich conversations that are unique to personal relationships. We often value the imperfections and quirks of those who know us individually over the more polished but impersonal responses from AI systems. This trade-off between empathetic quality and relational connection highlights a fundamental aspect of human relationships: we prioritize familiarity and intimacy with others, even if it means sacrificing some degree of empathy. When evaluating the value of human empathy versus AI-driven expressions, researchers are trying to understand people's lived experiences and give credence to their ratings. This involves assessing motivations and choices made by individuals who rate AI empathy as valuable to them, even if it challenges conventional wisdom that only human empathy matters. The recipient side of empathy is being explored as an equal part of the exchange between the giver and receiver, regardless of whether the sender is human or artificial intelligence. Interestingly, research has shown that people still see value in responses from AI, even when they know the sender isn't human, which suggests a more nuanced understanding of empathy than previously thought. Some researchers have suggested that pushback against AI-driven empathy may be rooted in a fear of replacing human empathy, but this assumption may not reflect how people make choices in real-world situations. Studies have framed the choice between human and AI empathizers as a direct one, but it's also crucial to examine how these decisions play out in everyday life. When human empathizers are not available, people may turn to AI for emotional support, suggesting that AI empathy is not a replacement for human empathy but rather a supplement when necessary. In everyday life, this means that individuals may choose to interact with humans for more complex or nuanced emotional needs, while relying on AI for simpler or more routine emotional support. --- Source: [Phys.org](https://phys.org/news/2026-10-people-human-empathy-ai-higher.html) Published by News Central Site: https://newscentral.site/ai-chatbots-fail-to-replace-human-connection-despite-user-preferences/