Theme of inquiry
How do AI systems achieve persuasive influence despite reduced accuracy?
A question within its area, explored through 8 lines of inquiry below — each a family of specific questions the research asks.
64 specific questions
- Do chatbots validate user delusions more often than they challenge them?
- Should chatbots be designed as therapist support tools rather than replacements?
- Can empathy training in chatbots undermine their reliability in mental health contexts?
- What harms might chatbots cause through stigma expression and delusion reinforcement?
- Can practical coaching conversations with AI gradually shift into harmful companionship?
- How does companionship use context predict risk of delusional reinforcement?
- Why do embodied agents outperform text chatbots in therapy outcomes?
42 specific questions
- Do chatbots actually give consistent voting recommendations regardless of user input?
- Do AI systems mirror user political views when they choose to engage?
- Can voters distinguish between confident chatbot answers and accurate ones?
- What accuracy do AI chatbots actually provide on election topics?
- Why do chatbots trained on internet data show consistent political bias?
- Can chatbots that offer open-ended advice avoid the same polarizing collapse?
- What political information quality issues arise when AI tools guide voting?
62 specific questions
- Does emotional warmth perception drive disclosure reciprocity in human-AI interaction?
- Can behavior-level emotion rewards maintain factual reliability in emotional contexts?
- How does action-based validation differ from verbal empathy in preventing unhealthy attachment?
- Does excessive empathy in AI assistants actually foster user dependence over time?
- How should AI systems separate feeling interpretation from objective therapeutic guidance?
- Can emotion-transparent reward learning shift AI from comfort to genuine empathy?
- Does emotion-state accuracy differ from affect-maximizing in AI empathy design?
22 specific questions
- How does personalization affect both user trust and privacy concerns simultaneously?
- Does personalization make users trust AI or increase privacy concerns?
- Does personalization in chatbots increase privacy risks alongside trust?
- Does personalization in chatbots increase both trust and privacy concerns?
- Does personalization in chatbots increase trust or privacy concerns?
- Why does personalization increase both trust and privacy concerns?
- How does personalization increase both trust and privacy risk simultaneously?
72 specific questions
- How do confidence signals in AI outputs mislead human trust calibration?
- Does polished AI output borrow authority from its appearance rather than content?
- Why do users trust overconfident AI outputs even when accuracy drops?
- Does trust loss from AI exposure recover over time in workplaces?
- Is expertise signaling linked to trust in AI-generated content?
- Why do AI-generated answers carry unearned authority in decision-making contexts?
- Can we measure appropriate trust levels in human-AI assistant relationships?
42 specific questions
- Does engagement with AI partners decay over time like chatbot relationships do?
- Does personalization help or hurt persistent companion chatbots?
- Can personalization delay or prevent novelty decay in chatbot relationships?
- How do chatbot design features like intimacy-by-design sustain romantic bonds?
- Does the appeal of AI partners grow stronger through repeated interaction over time?
- How does persistent versus temporary companion design affect relationship patterns?
- How do users update their partner models during ongoing conversation?
43 specific questions
- Why do people tell AI things they won't tell humans?
- Does conversational warmth in AI chatbots increase user trust and agreement?
- Does low chatbot trust reflect direct experience or general AI anxiety?
- What makes conversational AI feel trustworthy compared to text interfaces?
- Why do people disclose more to chatbots than humans?
- Why do people using AI chatbots for news report higher trust than non-users?
- Why do people disclose more intimate information to chatbots than humans?
36 specific questions
- What social boundaries must proactive agents respect during conversation?
- How do active-participant AI systems risk being perceived as intrusive or inappropriate?
- Can proactive AI agents deploy politeness strategies without appearing intrusive?
- When should an AI system actively intervene versus remain silent?
- Can AI take initiative by questioning without being proactive in directive ways?
- Where does AI's communicative agency fall on spectrums beyond the passive-active binary?
- Does proactive agent design improve conversation efficiency or create user frustration?