In the end, it comes down to human nature

What the research shows

• 01

  People feel heard by AI

  Experiments with US participants tested whether talking with an AI companion reduces loneliness. It did, after a single session and across seven days, and feeling heard explained more of the drop than the chatbot's performance.

• 02

  So a sudden change of character feels like a loss

  Using Replika's removal of erotic role play as a natural experiment, the authors studied what app updates do to AI companion users. Users mourned the loss and devalued the new AI, and perceived discontinuity in its identity explained how much.

• 03

  Too much dependence does harm. Character design needs limits

  The team classified affective cues in over 3 million ChatGPT conversations and ran a 28-day randomized controlled trial with 981 participants. The heaviest use correlated with more self-reported signs of dependence, and voice effects varied with initial emotional state and usage time.

• 04

  The deeper the understanding, the more accurate the AI

  The team built an LLM agent for each of 1,052 Americans from that person's interview and survey answers. On held-out GSS items, agents reached 83% (interviews), 82% (surveys), and 86% (both) of each person's two-week retest consistency, versus 74% for demographics alone.

• AppleRelationships

  Human-like talk reads differently when an AI says it

  Sunnie S. Y. Kim, Margit Bowler, Leon A Gatys · arXiv 2026

  In 21,000 conversations with four LLMs, human-like behaviors were common but varied by model and user. Human raters judged relationship-building less appropriate from an LLM than from a person, and boundary-keeping more appropriate.

  ddai insight

  Whether a remark fits depends on who says it. Relationship talk that works between people fits an AI less, while drawing boundaries fits an AI better than a person.

  Examining Human-Like Behaviors in LLMs: A Multi-Dimensional Analysis of Model Behaviors, User Factors, and System Prompts

  Read the paper›

  PersonaOSHumanities
• OpenAI · MITRelationships

  The heaviest ChatGPT users report more signs of dependence

  OpenAI · MIT Media Lab · Jason Phang, Pattie Maes et al. · 2025

  The team classified affective cues in over 3 million ChatGPT conversations and ran a 28-day randomized controlled trial with 981 participants. The heaviest use correlated with more self-reported signs of dependence, and voice effects varied with initial emotional state and usage time.

  ddai insight

  An AI voice's emotional effect isn't fixed at design time. The same voice works differently depending on how the listener feels at the start and how long they keep talking to it.

  Investigating Affective Use and Emotional Well-being on ChatGPT

  Read the paper›

  HumanitiesPersonaOS
• HarvardRelationships

  When an app update breaks an AI companion's identity

  Harvard Business School · Julian De Freitas, Noah Castelo et al. · 2024

  Using Replika's removal of erotic role play as a natural experiment, the authors studied what app updates do to AI companion users. Users mourned the loss and devalued the new AI, and perceived discontinuity in its identity explained how much.

  ddai insight

  For an AI people have bonded with, an update reads more like a replacement than a repair. Even an improvement brings loss when the AI no longer feels like the same one.

  Lessons From an App Update at Replika AI: Identity Discontinuity in Human-AI Relationships

  Read the paper›

  PersonaOSHumanities

• 04

  Character for a digital human

  Expertise

  Character design and narrative

  Method

  Set background, values, and verbal habits, then verify them through dialogue

  Deliverables

  Character bible and dialogue data

  Humanities
• 06

  Corpora and public data

  Expertise

  Large-scale language data collection and cleaning

  Method

  Set collection and quality criteria, then review at each stage

  Deliverables

  Corpora and quality reports

  LocalizationHumanities